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
|
|---|---|---|---|---|---|
Inflammation is a highly coordinated host response to infection , injury , or cell stress . In most instances , the inflammatory response is pro-survival and is aimed at restoring physiological tissue homeostasis and eliminating invading pathogens , although exuberant inflammation can lead to tissue damage and death . Intravascular injection of adenovirus ( Ad ) results in virus accumulation in resident tissue macrophages that trigger activation of CXCL1 and CXCL2 chemokines via the IL-1α-IL-1RI signaling pathway . However , the mechanistic role and functional significance of this pathway in orchestrating cellular inflammatory responses to the virus in vivo remain unclear . Resident metallophilic macrophages expressing macrophage receptor with collagenous structure ( MARCO+ ) in the splenic marginal zone ( MZ ) play the principal role in trapping Ad from the blood . Here we show that intravascular Ad administration leads to the rapid recruitment of Ly-6G+7/4+ polymorphonuclear leukocytes ( PMNs ) in the splenic MZ , the anatomical compartment that remains free of PMNs when these cells are purged from the bone marrow via a non-inflammatory stimulus . Furthermore , PMN recruitment in the splenic MZ resulted in elimination of virus-containing cells . IL-1α-IL-1RI signaling is only partially responsible for PMN recruitment in the MZ and requires CXCR2 , but not CXCR1 signaling . We further found reduced recruitment of PMNs in the splenic MZ in complement C3-deficient mice , and that pre-treatment of IL-1α-deficient , but not wild-type mice , with complement inhibitor CR2-Crry ( inhibits all complement pathways at C3 activation ) or CR2-fH ( inhibits only the alternative complement activation pathway ) prior to Ad infection , abrogates PMN recruitment to the MZ and prevents elimination of MARCO+ macrophages from the spleen . Collectively , our study reveals a non-redundant role of the molecular factors of innate immunity – the chemokine-activating IL-1α-IL-1RI-CXCR2 axis and complement – in orchestrating local inflammation and functional cooperation of PMNs and resident macrophages in the splenic MZ , which collectively contribute to limiting disseminated pathogen spread via elimination of virus-containing cells .
Viral vectors based on adenovirus of human and animal types have been widely adapted for gene transfer applications both in vitro and in vivo . Adenovirus ( Ad ) is remarkably efficient at infecting both dividing and non-dividing cells and its non-enveloped capsid tolerates major modifications that may restrict or target virus entry into desired cell or tissue types [1] . In permissive cell types , Ad replication follows the lytic cycle , which culminates in productive virus genome replication and release of progeny virions upon cell lysis . This feature of the virus is adopted in the concept of oncolytic Ad vectors , where through specific genetic modifications of the viral genome , virus replication and cell lysis are restricted to cancer cells [2] . Although in preclinical and clinical studies local delivery of oncolytic Ad was found to be safe , intravascular administration of Ad vectors , especially at high doses , activates host innate immune and inflammatory responses that may result in morbidity and mortality [3] , [4] , [5] , [6] . Molecular mechanisms responsible for the activation of severe innate immune and inflammatory responses to Ad remain poorly understood . Macrophages are the first line of cellular defense against invading pathogens . Tissue resident macrophages in the liver ( Kupffer cells ) , spleen ( macrophages expressing macrophage receptor with collagenous structure , MARCO+ , in the splenic marginal zone ) , and lung sequester Ad particles after their intravascular administration [7] , [8] , [9] , [10] . These cells are believed to be the principal activators of inflammation in response to Ad since in macrophage-depleted mice , Ad administration leads to greatly reduced inflammatory responses , compared to control un-manipulated animals [7] , [11] . Using gene deficient mice , we previously showed that inflammatory cytokines and chemokines were activated by Kupffer cells in the liver and MARCO+ macrophages in the spleen as early as 10 minutes after intravenous Ad administration [10] . We further showed that IL-1α-IL-1RI was a key pathway driving inflammatory cytokine and chemokine activation at this early time point after virus injection [10] . Despite our better understanding of the molecular mediators of inflammation triggered by Ad in vivo , their contribution to orchestrating cellular host inflammatory responses to the virus remains unclear . In this study , we analyzed the functional consequences of activation of IL-1α-IL-1RI-dependent and independent pathways in triggering cellular inflammatory responses to Ad in a model of acute disseminated infection in mice . We found that the IL-1α-IL-1RI-CXCR2 signaling axis cooperates with complement to recruit Ly-6G+7/4+ polymorphonuclear leukocytes ( PMNs ) to the splenic marginal zone in the proximity of virus-containing MARCO+ residential MZ macrophages . This PMN accumulation in the splenic MZ is associated with elimination of MARCO+ macrophages from the spleen . Pharmacological inhibition of complement activation in IL-1α-deficient mice prevents PMN recruitment to the splenic MZ and elimination of virus-containing macrophages . Our study reveals non-redundant roles of distinct molecular components of the innate immune response - IL-1α-IL-1RI-driven activation of CXCL1 and CXCL2 chemokines and complement - in enabling recruitment and functional cooperation of cellular components of innate immunity for elimination of virus-containing cells in the model of acute disseminated Ad infection .
Intravascular injection is a preferred delivery route for viral gene transfer vectors that are aimed at targeting specific disease-affected tissues or disseminated cancer cells in vivo . In clinical settings , disseminated Ad infections with virus titers reaching as high as 108 to 1010 virus particles per ml of blood can also be observed in a subset of immunocompromized patients undergoing bone marrow transplantation [12] , [13] , [14] , [15] , [16] . To analyze the acute cellular host responses to intravascular virus injection , we infected wild type mice with slightly lower amounts of HAd5-based vectors ( 5×109/ml of blood , which is equal to 1010 virus particles per mouse ) . This and higher virus doses are known to induce strong inflammatory cytokine production [3] , [17] , [18] . Two hours after virus injection , mice were sacrificed and the cellular compositions of bone marrow , blood , and spleen were analyzed by flow cytometry . Neutophils are the polymorphonuclear leukocytes of the myeloid lineage ( PMNs ) , and they are among the most abundant leukocytes in the peripheral blood and the bone marrow that promptly respond to tissue injury or infection by migrating to potential infection sites and deploying an array of effector factors that aim at limiting invading pathogens [19] . The analysis of the number of inflammatory PMNs ( that can be identified by staining with antibodies recognizing Ly-6G and 7/4 cell surface markers [20] ) in the bone marrow after saline ( Mock ) or Ad injection showed that there was a slight , but significant reduction in the absolute number of PMNs in the bone marrow after virus administration , compared to the control mock-injected group ( Fig . 1A–B ) . Considering that the number of total bone marrow cells in one femur bone of a mouse is equivalent to 6 . 7% of total marrow cells in the mouse body , the detected egress of 2×106 Ly-6G+7/4+ cells from the bone marrow harvested from one femur bone is equivalent to 30×106 PMN cells leaving the marrow and entering the bloodstream and peripheral sites in response to Ad administration . In agreement with this assumption , the analysis of the blood and spleen for the presence of Ly-6G+7/4+ PMNs showed that the absolute numbers of these cells were greatly increased in both peripheral blood and spleen after Ad injection ( Fig . 1 ) . Although it was previously demonstrated that PMNs enter peripheral tissues ( such as liver ) after intravascular Ad administration [17] , [21] , [22] , the quantitative analysis of leukocyte populations of the liver is technically challenging . In contrast , the methods to analyze cellular composition of the spleen are well established , therefore reliable quantitative data can be obtained by using both immuno-fluorescence and flow cytometry approaches . Using flow cytometry analysis , we found that the number of Ly-6G+7/4+ cells in the spleen increased 5-fold , compared to saline-injected animals , and reached the absolute number of 10×106 cells after injection of mice with Ad ( Fig . 1E–F ) . This data indicates that at the time of analysis ( 2 hours after the virus injection ) , one-third of PMNs that have been released from the bone marrow , were detectable in the peripheral blood , one-third in the spleen , and the remaining cells are likely to be distributed among other peripheral tissues . To analyze whether accumulation of PMNs in the spleen was virus-dose- and/or time-dependent , we injected mice with Ad at doses ranging from 109 to 1011 virus particles ( vp ) per mouse and harvested spleens for flow cytometry analyses from 0 hours to 8 hours after virus administration . This analysis showed that when Ad was injected into mice at doses ranging from 109 to 1010 vp per mouse , the number of PMNs in the spleen increased rapidly , reaching its maximum at 2 hours post virus injection , and then steadily declined by 8 hours after the virus administration ( Fig . 2 ) . At a virus dose of 1011 vp per mouse , the number of PMNs continued to increase at all time points analyzed and at 8 hours after virus injection the proportion of PMNs in the spleen of mice injected with this dose of virus was four-fold greater than mice administered with 109 vp and two-fold greater than in mice administered with 1010 vp/mouse . Taken together , these analyses revealed that in response to intravascular Ad administration , PMNs with a Ly-6G+7/4+ pro-inflammatory phenotype are rapidly released from the bone marrow into the blood . The accumulation of PMNs in the spleen was virus dose dependent , with the highest number of cells observed at 2 hours after the virus injection for the doses of up to 1010 vp/mouse . At this time point , one-third of the total Ly-6G+7/4+ cells released from the marrow was recovered from the spleen . The spleen is the largest secondary immune organ in the body and is responsible for initiating immune responses to blood-borne antigens and for filtering the blood of old and damaged red blood cells . The spleen is comprised of two functionally and anatomically distinct compartments , the red pulp and the white pulp . The red pulp is a blood filter that removes foreign material and damaged erythrocytes . The white pulp is composed of three subcompartments: the periarteriolar lymphoid sheath ( PALS ) , the follicles , and the marginal zone ( MZ ) [23] , [24] , [25] . Blood flowing through the marginal sinus and marginal zone percolates through the marginal zone in the direction of the red pulp . Previously , we had shown that after intravascular administration , Ad particles are sequestered in the spleen specifically by the MARCO+ marginal zone macrophages [10] . To analyze whether Ly-6G+7/4+ PMNs localize to a particular anatomical compartment in the spleen after Ad injection , we injected mice with Ad and analyzed distribution of PMNs on spleen sections 2 hours after virus injection . In agreement with previous findings [10] , staining of spleen sections with anti-Ad hexon antibody demonstrated that the vast majority of the hexon-specific staining localized to the splenic marginal zone ( Fig . 3A ) . Staining of the spleen sections with Ly-6G- , Gr-1- , or 7/4 PMN marker-specific antibodies revealed identical patterns of distribution of positive cells , and the vast majority of these cells were also localized to the marginal zone , and not to other splenic anatomical compartments ( Fig . 3A and B ) . It is worth noting that when mice were injected with saline , less than 5% of PMNs were localized to the splenic MZ ( Fig . 3A , Mock group ) , suggesting that PMN recruitment in the MZ may constitute a spleen-specific host response to Ad administration . To further delineate whether PMN localization to the MZ after Ad administration is specific and not merely a reflection of the increased number of circulating Ly-6G+7/4+ cells in the peripheral blood , we purged monocytes from the bone morrow via a non-inflammatory stimulus by using the small drug AMD3100 that interferes with SDF1-CXCR4-mediated retention of PMNs and stem/progenitor cells in the bone marrow [26] , [27] , [28] , [29] . In agreement with previous findings , administration of AMD3100 resulted in a massive release of mature leukocytes , including Ly-6G+ cells , from the bone marrow into the circulation and peripheral tissues ( Fig . 3C ) . However , the accumulation of these cells in the splenic MZ was not different from that observed in saline-injected mice , with less than 8% localizing in MZ ( Fig . 3D ) . These data provide direct evidence that under local non-inflammatory conditions , the traffic of Ly-6G+ cells through the spleen does not result in their accumulation in the MZ . These data further suggest that entry and retention of PMNs in the splenic MZ in response to Ad is specific and is not merely a reflection of passive accumulation in the spleen from a circulating granulocyte pool in the blood observed after virus administration . Because release of granulocytes from the bone marrow mediated by AMD3100 is non-inflammatory [30] , we hypothesized that the recruitment of PMNs in the splenic MZ after Ad administration occurs in response to local inflammatory stimuli released by MZ macrophages that trap virus particles . To test this hypothesis , we analyzed a panel of inflammatory cytokines and chemokines in the spleen after virus administration using a protein-profiler immuno array [10] . Using this method of analysis we earlier showed that the intensity of dots measured in histogram units directly correlates with the absolute amount of the analyte under investigation [31] . This analysis confirmed that the amounts of inflammatory cytokines and chemokines in the spleen were highly elevated after injection of unmodified HAdv5 and not after injection of viruses that are also sequestered by the MZ cells but are known to induce low-level inflammatory cytokine production ( Fig . 4A–B ) . Ad5/35S is an HAdv5-based vector that was previously shown to poorly activate inflammatory cytokine production due to a mutation in the virus fiber protein [32] . The Ad-ts1 is a thermo-sensitive mutant of HAdv2 and cannot escape the endosomal cellular compartment after internalization into the cell [33] , [34] . This virus is also known to induce a low-level inflammatory response after intravascular administration [10] . Using flow cytometry analysis , we next found that the number of Ly-6G+7/4+ cells in the spleen significantly increased only after injection of mice with HAdv5 , but not with Ad5/35S or Ad-ts1 viruses ( Fig . 4C ) . Importantly , the distribution of PMNs in the spleens of mice injected with Ad5/35S and Ad-ts1 viruses was significantly different from that observed in mice injected with HAdv5 , where the vast majority of Ly-6G+ cells was found in the splenic MZ ( Fig . 4D–E ) . In contrast , after mouse injection with Ad5/35 and Ad-ts1 viruses , the majority of PMNs was found in the red pulp and not in the MZ , despite the evident accumulation of virus particles in MZ cells . This distribution of PMNs closely resembles their localization in Mock-injected group , where mice were administered with saline ( Fig . 4D–E ) . Collectively , our analyses showed that in response to intravascular Ad administration , after release from the bone marrow , PMNs enter into and are retained in the splenic MZ . PMNs that are released from the bone marrow via a non-inflammatory stimulus do not enter or become retained within the MZ compartment of the spleen . The Ad5/35S and Ad-ts1 vectors , which induce low-level inflammatory cytokine and chemokine activation , trigger low levels of entry and retention of PMNs in the MZ , despite the efficient accumulation of virus particles in MZ cells ( Fig . 4E ) . We previously showed that upon entry into macrophages in vivo , Ad activates a stereotypic cascade of inflammatory cytokines and chemokines through activation of IL-1α [10] . IL-1α is a principal pro-inflammatory cytokine produced by the vast majority of cell types in the context of necrotic cell death [35] , [36] , [37] . Similar to IL-1β , IL-1α binds to IL-1 receptor type 1 ( IL-1RI ) , and both of these cytokines activate identical biological responses downstream of IL-1RI signaling [38] , [39] , which include inducing CXCL1 and CXCL2 chemokines , which trigger PMN recruitment . To analyze whether IL-1RI signaling or individual IL-1RI ligands were responsible for PMN retention in the splenic MZ , we first administered Ad into wild type ( WT ) mice and mice deficient in IL-1α , IL-1β , IL-1α/β , and IL-1RI , and analyzed the percentage of Ly-6G+7/4+ cells in the spleen 2 hours after virus administration . This analysis revealed that the number of Ly-6G+7/4+ cells was significantly higher in the spleens of mice injected with Ad , compared to mock-injected group . However , the number of PMNs in the spleens of mice after Ad injection was two-fold higher in WT and Il1b−/− mice , compared to Il1a−/− , Il1a/b−/− , and Il1r1−/− animals ( Fig . 5A ) . These data suggest that the lack of IL-1α-IL-1RI signaling may either slightly inhibit the egress of PMNs from the bone marrow or selectively prevent their entry into the spleen . Importantly , the analysis of distribution of PMNs between the red pulp and MZ revealed that while in WT mice , over 80% of PMNs localized within the MZ compartment , in both Il1a−/− and Il1a/b−/− animals , the PMN distribution was significantly different from that observed in both the virus-injected and saline-injected WT mice . Specifically , in Il1a−/− and Il1a/b−/− mice , on average , 60% of cells localized to red pulp and 40% localized to the MZ ( Fig . 5B–C ) . Although the proportion of PMNs localizing to the MZ in Il1a−/− and Il1a/b−/− animals was significantly reduced compared to the Ad-injected WT mice , this number was still significantly higher , compared to that observed in saline-injected animal where only 5–7% of PMNs were found in the splenic MZ . These data indicate that IL-1α-IL-1RI signaling is only partially responsible for guiding the recruitment of PMNs into the splenic MZ after Ad injection . PMN recruitment to the sites of infection occurs in response to chemotactic stimuli , with CXCL1 and CXCL2 being among the most potent chemokines promoting PMN migration [19] . However , both of these chemokines can bind to CXCR1 and CXCR2 receptors on neutrophils to promote their migration in various pathological conditions [19] , [40] , [41] . To further delineate which chemokine receptor plays a role in guiding PMN recruitment into the splenic MZ after intravascular Ad injection , we administered virus to WT and Cxcr1−/− or Cxcr2−/− mice . The analysis of the percentages of Ly-6G+7/4+ cells in the spleen showed that in both WT and Cxcr1−/− mice , the numbers of PMNs in the spleen were identical 2 hours after the virus injection . Although the number of PMNs in the spleens of Cxcr2−/− mice was significantly higher than found in saline-injected WT animals ( Mock group , Fig . 6A ) , it was two-fold lower , compared to the percentage of PMNs recovered from the spleens of virus-injected WT and Cxcr1−/− animals . The analysis of PMN distribution between the red pulp and MZ after injection of mice with Ad further showed that the distribution of Ly-6G+ cells in splenic parenchyma was identical in WT and Cxcr1−/− mice , with over 80% of cells localizing in the splenic MZ in both variants ( Fig . 6B–C ) . In contrast , in Cxcr2−/− mice , only 30–35% of PMNs localized to the splenic MZ , and the majority of Ly-6G+ cells localized to red pulp after Ad administration . These data demonstrate that efficient PMN recruitment and/or retention in the splenic MZ requires functional CXCR2 , and not CXCR1 , signaling . It is also consistent with , and qualitatively and quantitatively phenocopies , the PMN distribution in Il1a−/− , Il1a/b−/− , and Il1r1−/− mice , and suggests that CXCL1 and CXCL2 chemokines ( Fig . 4A ) , which are activated downstream of IL-1α-IL-1RI signaling in response to intravenous Ad injection , signal through the CXCR2 receptor on PMNs to mediate their recruitment and retention in the splenic MZ , the anatomic compartment that contains macrophage populations sequestering Ad from the blood . After intravascular administration , Ad particles are trapped in the residential MARCO+ macrophages in the splenic MZ ( Fig . 7A and [10] ) . We next analyzed the kinetics of changes in the MARCO+ cell population in the spleen after virus administration over time . Using immuno-histochemical staining of spleen sections with anti-MARCO-specific antibodies , we found that the MARCO+ MZ cells progressively disappear , and by 24 hours after virus injection , the cellularity of MZ was evidently reduced , compared to control , saline-injected groups ( Fig . 7B–C ) . The disappearance of MARCO+ cells from the splenic MZ was dose-dependent and occurred at virus doses of 1010 per mouse and higher ( Figure S1 ) . To distinguish between the possibilities that MARCO+ cells might have migrated out of the splenic MZ to other host compartments or died in situ , similarly to CD68+ residential macrophages in the liver [42] , we conducted an evaluation of ultra-structural changes within the MZ cells that contain Ad particles using conventional transmission electron microscopy . This analysis revealed that unlike MZ macrophages in saline-injected animals , virus-containing MZ macrophages in Ad-injected mice were filled with abnormal vacuolized cytosolic compartments and grossly-distorted mitochondria ( Fig . 7D–G ) . Therefore , the virus-containing cells in the splenic MZ are unlikely to maintain their functional physiological state , but instead are undergoing a catastrophic disorganization of the cytosolic compartments that is consistent with initiation and/or execution of a cell death program in situ . Depending on physiological or patho-physiological stimuli , cells can initiate and execute various cell demise programs [43] , [44] . We recently found that upon sequestering Ad particles from the blood , CD68+ residential macrophages in the liver ( Kupffer cells ) initiate and execute a unique necrotic-type cell death pathway that depends on the transcription factor IRF3 [42] . This type of cell death was associated with the major disorganization of the cytosolic compartments and induction of plasma membrane permeability to propidium iodide ( PI ) , which can be detected in situ via analyzing PI-stained nuclei in the livers of virus-injected mice [42] . To define whether MARCO+ macrophages in the spleen also undergo necrotic-type cell death after interaction with Ad , we administered Ad and PI to mice and analyzed the distribution of PI+ cells in sections of liver and spleen 1 hour after virus injection . Consistent with previous observations , this analysis showed that after injection of mice with Ad , the liver parenchyma contained up to 150 PI+ cells per view field . However , we failed to observe PI+ cells in the spleen at the same time point ( Fig . 8A ) , suggesting that MZ macrophages do not undergo necrotic-type cell death after interaction with Ad . In agreement with earlier studies [42] , [45] , we further found that the Ad-ts1 and Ad5/35S virus vectors ( that induce inflammatory response poorly ( Fig . 4 ) ) , do not trigger PI plasma membrane permeability in the spleen or liver ( Fig . 8A ) . To definitively exclude that MZ macrophages undergo IRF3-dependent necrosis after interaction with adenovirus , we administered Ad into WT and Irf3−/− mice and analyzed persistence of the MARCO+ cell population over time . This analysis showed that at 1 hour after the virus injection , MARCO+ macrophages efficiently trapped virus particles from the blood ( Fig . 8B ) . However , by 24 hours after virus injection , MARCO+ cells were missing in both the WT and Irf3−/− mice ( Fig . 8C–D ) . In contrast , administration of WT mice with Ad-ts1 virus showed that this macrophage population remained in the MZ 24 hours after the virus injection irrespective of the efficient accumulation of virus particles in MZ cells ( Fig . 4E ) . Collectively , this data shows that resident macrophages in the liver and spleen respond to Ad particles via distinct pathways . Although after sequestering Ad particles from the blood , both liver Kupffer cells [42] , [45] , [46] and MZ macrophages undergo catastrophic disorganization of cytosolic compartments , MZ macrophages do not lose their plasma membrane integrity and are eliminated from the spleen in both WT and Irf3−/− mice . The finding that MZ macrophages do not lose plasma membrane integrity after sequestering Ad from the blood was surprising and reveals a clear divergence in macrophage responses to Ad in a tissue-specific manner . Furthermore , although after intravascular Ad injection PMNs do enter the liver [21] , [22] , they are unlikely to directly contribute to elimination of virus-containing Kupffer cells , since these cells undergo IRF3-dependent necrosis in a cell-autonomous fashion [42] . In contrast , the recruitment and retention of PMNs in the splenic MZ , that is the natural host compartment for virus-trapping MARCO+ macrophages , may indicate that these PMNs have a functional role in eliminating virus-containing cells from the MZ . To investigate this possibility , we first analyzed the localization of Ly-6G+ cells in the splenic MZ of Il1a/b−/− mice in relation to localization of the Ad-containing cells . This analysis revealed that when IL-1-IL-1RI signaling is lacking , those PMNs that enter and are retained in the MZ localize in the immediate proximity to the virus-containing cells ( Fig . 9A ) . This characteristic PMN localization in the immediate proximity to the virus-containing MZ cells suggests that even without IL-1R ligands , there are additional , locally-presented chemotactic stimuli that guide PMN migration to the MZ and position them in contact with Ad-containing cells . Besides CXC chemokines , CXCL1 and CXCL2 , the activated components of a complement cascade , e . g . C3a and C5a , were shown to be potent chemoattractants for PMN localization to the sites of infection [40] . Furthermore , extensive earlier studies showed that complement is indeed activated in response to intravascular Ad administration in vivo and C3a fragments are readily detectable in the blood of virus-injected mice [47] . Of note , the Ad-ts1 virus that fails to initiate IL-1RI-dependent chemokine activation ( [10] and Fig . 4A ) , PMN recruitment to the MZ ( Fig . 4E ) , and MARCO+ cell death ( Fig . 8E ) , also failed to trigger complement activation in vivo [47] . To analyze whether complement may play a role in PMN recruitment to the splenic MZ after Ad injection , we administered the virus to WT and complement component C3-deficient ( C3−/− ) mice . The analysis of PMN distribution on the spleen sections of virus-injected mice showed that the proportion of PMNs that localized to the MZ after Ad injection in C3−/− mice was significantly lower than in WT mice ( Fig . 9B–C ) . While over 80% of PMNs localized to the splenic MZ in WT mice , only 22% of PMNs localized to the MZ in C3−/− mice . Accordingly , the majority of PMNs in virus-injected C3−/− mice were distributed throughout the red pulp of the spleen . We further confirmed that C3−/− mice were not deficient in initiation of IL-1α-IL-1RI signaling and production of CXCL1 and CXCL2 chemokines ( Fig . 9D ) . We observed a reduced expression of IL-6 in C3−/− mice after virus injection as compared to WT mice . However , the Il6−/− mice exhibited no defects in activating IL-1α-IL-1RI-signaling and CXCL1 and CXCL2 chemokine expression or recruiting PMNs to the splenic MZ after virus administration ( data not shown ) . To analyze whether IL-1α-IL-1RI signaling and complement cooperate in PMN recruitment to the splenic MZ after intravascular Ad injection , we utilized a pharmacological approach of blocking all or only alternative complement activation pathways by pre-treating mice with CR2-Crry or CR2-fH complement inhibitory proteins [48] prior to Ad administration . Pre-treatment of WT mice with CR2-Crry [49] or CR2-fH [50] prior to Ad injection resulted in partial re-distribution of PMNs away from the MZ ( Fig . 10A–D , WT panels ) . However , pre-treatment of IL-1α-deficient mice with complement-blocking proteins resulted in complete inhibition of PMN localization to the MZ , despite evident accumulation of virus particles in MZ cells ( Fig . 10A–D , Il1a−/− panels ) . The analysis of MARCO+ cell populations in WT and Il1a−/− mice after Ad injection with pre-treatment of mice with CR2-Crry ( not shown ) or CR2-fH showed that the pre-treatment of WT mice with CR2-fH prior to Ad injection did not prevent elimination of MARCO+ cells from the MZ 24 hours after the virus administration . In contrast , pre-treatment of mice deficient in IL-1α with CR2-fH prior to the virus administration resulted not only in preservation of MARCO+ cells in the MZ , but also in appearance of MARCO+ cells in the red pulp – a response that was not observed in CR2-fH-treated WT mice ( Fig . 10E–F ) . Taken together , these analyses showed that the recruitment of PMNs to the MZ after Ad injection is deficient in complement component C3−/− mice . They also showed that complement and IL-1α-dependent signaling cooperate in PMN recruitment into the MZ . Pharmacological inhibition of complement in IL-1α-deficient mice prevented PMN accumulation in the splenic MZ and elimination of MARCO+ virus-containing cells .
Adenovirus has been adopted as a highly efficient gene transfer vehicle to deliver exogenous genes to various cell types and tissues in vivo . However , when the virus is injected intravenously , it induces acute and potent innate immune and inflammatory responses that can be lethal to the host [3] , [4] , [5] . Although numerous events , including cytokine and chemokine gene activation [17] , [18] , complement activation [51] , [52] , release of platelet activating factor [53] , and Kupffer cell death [42] , [46] were shown to ensue upon intravascular Ad administration , the specific contribution of each of these factors in activating systemic host inflammatory responses and their functional consequences remain poorly defined . Furthermore , the vast majority of studies conducted to date were focused largely on defining the most prominent molecular mediators that lead to acute systemic Ad-induced toxicity . Using a panel of genetic mouse models that are deficient in key mediators of innate immunity and inflammation , we previously found that Ad induces activation of a stereotypic pro-inflammatory cytokine and chemokine cascade through the IL-1α-IL-1RI signaling pathway [10] . However , the relevance of this pathway to the induction of cellular innate responses remains unclear . Here we analyzed host cellular inflammatory response to Ad in wild type mice and mice deficient in IL-1α-IL-1RI signaling pathway genes , CXCR1 and CXCR2 chemokine receptors , and complement . Our studies revealed that intravascular Ad administration leads to a release of polymorphonuclear leukocytes with pro-inflammatory Ly-6G and 7/4 cellular markers from the bone marrow into the blood and their entry and retention in the peripheral tissues , such as spleen . Previous studies showed that in response to intravascular Ad injection , no leukocyte retention was seen in the sinusoids of the liver [21] . In contrast to this observation , we found that PMN recruitment and retention to the spleen was a highly coordinated process that leads to redistribution of recruited cells to a confined and specific anatomical compartment of the spleen , the marginal zone . This area of the spleen is occupied by MZ macrophages that express the MARCO cell surface marker and selectively trap Ad particles after intravascular virus administration [10] . Trafficking of blood-borne material and cells through the spleen is highly complex and the exact routes of blood flow through the splenic parenchyma remain a debatable issue [23] . To define the exact natural trafficking route of Ly-6G+7/4+ PMNs through the spleen without challenging animals with Ad , we injected mice with the small molecule drug AMD3100 . Injection of mice with AMD3100 did not induce inflammatory cytokines and chemokine in the spleen ( data not shown ) and over 95% of Ly-6G+ PMNs were found in the splenic red pulp , outside of the MZ ( Fig . 3 ) . In contrast , after Ad administration , Ly-6G+ PMNs localized virtually exclusively to the splenic MZ , indicating the existence of specific and local stimuli that trigger PMN recruitment and retention in this defined anatomical compartment . CXC family chemokines , CXCL1 and CXCL2 , are among the most potent known chemoattractants that promote neutrophil migration and activation [19] . In response to intravascular Ad injection , both CXCL1 and CXCL2 are produced in the spleen downstream of IL-1α-IL-1RI signaling , a process that is initiated by MZ macrophages that trap Ad particles from the blood [10] . Here we found that in mice deficient in IL-1α , IL-1α/β , or IL-1RI , the recruitment of PMNs to the splenic MZ is reduced only partially , suggesting that additional stimuli may exist that guide PMN migration to and/or retention in the MZ . We also found that this phenotype was reproduced in CXCR2-deficient mice , but not in mice lacking CXCR1 . We further serendipitously found that in mice lacking complement component C3 , the recruitment of PMNs to the MZ after Ad injection was greatly reduced compared to the WT mice . Although complement is known to be activated after intravascular Ad delivery , the role of complement in driving neutrophil recruitment to the compartments in peripheral tissues that house virus-containing cells has never been demonstrated . Importantly , pharmacological blocking of complement activation in IL-1α-deficient , but not in WT , mice completely prevented PMN recruitment into the MZ despite the efficient trapping of virus particles by MZ cells . These data indicate that IL-1α-dependent signaling and complement cooperate in PMN recruitment to the splenic MZ and positioning of these cells in proximity to virus-containing macrophages . Neutrophils are among the most abundant professional phagocytic cells of the myeloid lineage and , together with circulating monocytes and tissue macrophages , form the effector arm of innate immunity [41] . Although both PMNs and macrophages can efficiently engulf and destroy extracellular pathogens , their effector mechanisms are distinct and non-redundant . Furthermore , macrophages and PMNs are known to cooperate at mounting host protective immune responses to both intracellular and extracellular bacterial pathogens [41] via diverse mechanisms . Phagocytosis , degranulation , and release of nuclear DNA fragments ( NETs ) are thought to be the principal PMN effector mechanisms enabling killing of pathogens [19] . However , it was also reported that the cooperation between macrophages ( that may harbor the pathogens ) and PMNs can occur through the transfer of anti-microbial molecules from PMNs to macrophages during degranulation of activated neutrophils [54] , [55] . While components of neutrophilic granules can kill bacterial pathogens , they also can exert cytotoxic activity and cause rapid death of surrounding cells [56] . In this context , we found that the physiological consequence of the PMN migration and retention in the splenic MZ and their localization in proximity to MARCO+ macrophages is elimination of the virus-containing cells from the MZ . The Ad-ts1 virus , which causes no PMN retention in the MZ , failed to trigger elimination of virus-containing MARCO+ cells . Furthermore , pharmacological inhibition of complement in IL-1α-deficient mice resulted in re-distribution of PMNs from the MZ to red pulp and preservation of the MARCO+ cell population despite effective retention of virus particles . It is noteworthy that in our recent study we found that Tlr4−/− mice also exhibited reduced PMN recruitment into the splenic MZ and failed to clear the virus from the spleen , when analyzed for the amounts of Ad DNA present 24 hours after virus infection by real-time PCR ( [57] , Figs . S15–S16 ) . Ad administration via an intravascular route causes multifaceted systemic innate immune and inflammatory responses . Numerous comprehensive analyses of virus bio-distribution after intravascular injection have shown that the liver and spleen accumulate over 90 percent of the administered virus particles [58] , [59] , [60] , [61] , [62] . Our study revealed that although the functional consequence of Ad interaction with tissue macrophages is virus elimination in both the liver and spleen , the molecular mechanisms employed by the host to restrict systemic virus spread are distinctly different ( Fig . 11 ) . While liver residential macrophages , Kupffer cells , undergo cell-autonomous IRF3-dependent necrosis after trapping Ad particles from the blood [42] , MARCO+ MZ macrophages in the spleen do not undergo cell-autonomous necrosis , but rather activate IL-1α-IL1RI-dependent pro-inflammmatory cytokine and chemokine production [10] . The interaction of Ad with splenic macrophages also triggers complement activation [47] and requires complement component C3 , which cooperates with CXCL1 and CXCL2 chemokines to recruit and retain PMNs with a pro-inflammatory phenotype ( Ly-6G+/7/4+ ) into the splenic MZ in contact with virus-containing cells . Activated PMNs degranulate locally and damage both the virus and virus-containing cells , leading to their demise and elimination from the spleen . Collectively , cooperation of tissue macrophages , which sequester Ad from the blood , and PMNs results in effective reduction of the virus load in the spleen . This model provides further support to the concept of molecular and functional specialization of tissue residential macrophages , that has been recently revealed at the molecular level through transcriptome profiling and functional characterization of macrophages from different organs [63] , [64] . Although the mouse is a rather resistant model to analyze systemic inflammation after Ad injection compared to rats [65] , non-human primates [3] and humans [5] , [6] , some clinical aspects of virus-induced systemic toxicity are well reproduced in mice . One of the clinically relevant signs of systemic Ad-induced toxicity observed is thrombocytopenia . Intravascular administration of Ads was shown to trigger severe thrombocytopenia in all animal models analyzed [3] , [4] , [66] , [67] . Remarkably , earlier studies showed that thrombocytopenia does not ensue after administration of Ad into spleenectomized mice [66] , suggesting that virus interaction with cells in the spleen may be mechanistically linked to triggering this deleterious host response . Prolonged retention of PMNs in tissues leads to collateral tissue damage due to local neutrophil degranulation and release of cell-damaging proteases , thus exacerbating inflammation and contributing to the organ failure . Our findings might prove helpful for researchers developing therapeutic protocols for translational use of gene transfer vectors based on Ad . Our finding that a combination of the pharmacological inhibition of complement activation and IL-1RI-signaling deficiency reduces PMN recruitment to the site of infection and spares the virus-containing cell population provides the rationale for reducing damage to host cells after intravascular administration of adenovirus vectors by using pharmacological intervention . Specifically , the combination of complement inhibitors [50] with clinically-approved inhibitors of IL-1RI signaling , e . g . recombinant IL-1RA “Anakinra” [68] , [69] , may prove useful for both modulating systemic toxicity of Ad after its intravascular administration as well as under other conditions where initial pro-inflammatory stimuli are produced by tissue residential macrophages that recruit activated PMNs and cause collateral tissue damage .
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 Washington , Seattle , WA , protocol number 4148-01 , and all efforts were made to minimize suffering . C57BL/6 mice were purchased from Charles River , Wilmington , MA . Cxcr1−/− , Cxcr2−/− , C3−/− , Il6−/− , and Il1r1−/− mice were purchased from Jackson Laboratory . Il1a−/− , Il1a−/−b−/− mice were described in [70] . Il1b−/− mice were described in [71] . All mice were on C57BL/6 genetic background , matched by age and housed in specific-pathogen-free facilities . The replication-defective Ad5-based vectors , HAdv5 ( Ad5L ) , Ad5/35S , and Ad-ts1 HAd2 mutant were previously constructed and described in detail elsewhere [8] , [32] , [72] , [73] . For Ad amplification , 293 cells were infected under conditions that prevented cross-contamination . Viruses were banded in CsCl gradients , dialyzed and stored in aliquots as described earlier [73] . Ad genome titers were determined by OD260 measurement . For each Ad used in this study , at least two independently prepared virus stocks were obtained . Each produced virus stock was tasted for endotoxin contamination using Limulus amebocyte lysate Pyrotell ( Cape Cod Inc , Falmouth , MA ) . For in vivo experiments , only virus preparations confirmed to be free of endotoxin contamination were used . Unless otherwise specified , mice were injected with 1010 virus particles in 200 µl of phosphate buffered saline ( PBS ) via tail vein infusion . At indicated times , mice were sacrificed and organs were harvested for further analyses . Previously published data shows that the frequency of marginal zone macrophages in the mouse spleen is 5 . 1% of splenocytes [74] . According to our earlier data on the absolute amount of Ad particles trapped in the spleen [10] , we estimated that if mice are injected with a dose of 1010 virus particles , each marginal zone macrophage will receive between 50 and 100 virus particles per cell . We experimentally determined that an average mouse spleen possesses 2 . 5×108 nucleated cells and yields on average 1 mg of total genomic DNA . Considering that the cumulative number of MZMφ is around 5% [74] , the total average number of MZMφ cells in a mouse spleen is approximately 1 . 25×107 . Our quantitative real-time PCR analysis shows that upon the injection of mice with Ad at a dose of 1010 virus particles per mouse , the total number of Ad genomes accumulated in the liver 30 min after virus injection is 7×105 per µg of splenic DNA , or 7×108 per whole spleen . By dividing 7×108 ( the number of Ad genomes per whole spleen ) by 1 . 25×107 ( the average number of MZMφ ) we find a maximum actual dose of around 50 Ad particles per each MZMφ cell . A “Proteome Profiler antibody array: Mouse Cytokine Array Panel A” ( #ARY006 , R&D System ) was used , according to the manufacturer's instructions . Each spleen was homogenized in 2 ml of sample solution , and 1 ml ( 1/2 spleen ) was used to incubate with each membrane on a rocking platform overnight [31] . Membranes were developed with ImmunoStar HRP-sustrate ( BioRad , #1705041 ) . Propidium iodide was purchased from Sigma-Aldrich , ( St . Louis , MO USA ) , Cat . #81845 , antibodies from Abcam: biotinylated anti-Ad-Hexon ( #ab34374 , final dilution 1/100 ) , anti-Ad5 ( #ab6982 , final dilution 1/50 ) ; Antibodies from BMA: anti-MARCO ( BMA , #T2026 , 2 ug/ml ) , Antibodies from BD: anti-IgM ( #553405 , 1 ug/ml ) , FITC-labeled anti-GR-1 ( #553127 ) , FITC-labeled anti-Ly-6G ( #551460 ) ; R-PE-labeled anti-7/4 ( Serotec MCA771PE Clone 7/4 ) . Antibodies against Secondary antibodies and reagents were from Jackson Immunoresearch: Cy2 or Cy3-labeled streptavidin , or donkey anti-rat or rabbit antibodies , Cy2- , Cy3- or HRP-labeled . AMD3100 compound was purchased from Sigma-Aldrich and suspended in phosphate–buffered saline ( PBS ) /bovine serum albumin ( BSA ) , and injected intraperitoneally at 50 mg per kg of mouse body weight . Complement inhibitor proteins CR2-Crry [49] and CR2-fH [50] were prepared as previously described . Splenocytes , bone marrow mononuclear cells , and peripheral blood were harvested from mice at indicated time points after adenovirus administration and stained by incubation with FITC-labeled anti-Ly-6G and PE-labeled anti-7/4 primary monoclonal antibodies for 30 minutes at 4°C in triplicate . Next , cells were washed and analyzed by fluorescence-activated cell sorting on FACSCalibur machine ( Becton Dickinson , San Jose , CA ) . Prior to flow cytometry analysis , red blood cells were lysed in all samples and nucleated cells were pelleted , washed , and processed as described above . Mice were anaesthetized , and spleens and livers were collected , frozen in O . C . T . compound and stored at −80°C until processed . Five consecutive 6–8 µm sections at 4 depth levels in the spleen and liver were cut , air dried , fixed for 10 minutes in acetone at −20°C , air dried for at least 4 hours , re-hydrated in TBS for one hour , blocked in 2% N . S . for 1 hour and incubated with primary antibodies overnight at 4°C with or without 0 . 1% saponin depending on the antigen . Then , sections were incubated with HRP-labeled secondary antibodies for 1 hour . Slides were developed with ImmPact DAB or NovaRed substrates ( Vector Laboratories ) , air dried , mounted , and analyzed on a Leica microscope . For immunofluorescence staining , slides were immediately mounted after washing with the secondary antibodies . Images were taken using a CCD camera-equipped Leica dual light fluorescent microscope . Four representative images were taken for each section cut from tissues at at least 3 depth levels for quantitative processing using MetaMorph 7 . 8 . 1 . 0 software under interactive threshold settings to define positive staining on the image and applied universally to all images in analyzed experimental groups . Statistical analysis in each independent experiment was performed with using one-way ANOVA followed by Newman-Kleus post-hoc test with GraphPad Prism 5 . 02 software . Data are reported as mean ± standard deviation . P<0 . 05 was considered statistically significant . All accession numbers are provided as shown in Swiss-Prot database: mouse interleukin 1 alpha ( IL-1β ) accession number P01582; mouse interleukin 1 beta ( IL-1β ) accession number P10749; mouse interleukin 1 receptor type I ( IL-1RI ) accession number P13504; mouse complement component C3 ( C3 ) accession number P01027; mouse growth regulated alpha protein ( CXCL1 ) accession number P12850; mouse C-X-C chemokine 2 ( CXCL2 ) accession number P10889; mouse C-X-C chemokine receptor type 1 ( CXCR1 ) accession number Q810W6; mouse C-X-C chemokine receptor type 2 ( CXCR2 ) accession number P35343; mouse interleukin 6 ( IL-6 ) accession number P08505 . | Adenovirus ( Ad ) induces a potent activation of pro-inflammatory cytokines and chemokines upon interaction with tissue macrophages in vivo . However , critical factors affecting cellular inflammatory responses to Ad and their functional significance remain unclear . Here we show that in the model of disseminated infection , intravenous Ad administration leads to a rapid release of pro-inflammatory Ly-6G+7/4+ leukocytes ( PMNs ) from the bone marrow into the blood . PMNs enter into peripheral tissues and , in the case of spleen , are accumulated in proximity to the virus-containing MARCO+ macrophages within the splenic marginal zone ( MZ ) . Mechanistic dissection of molecular queues that guide PMN migration reveals that CXCL1 and CXCL2 chemokines are only partially responsible for CXCR2-dependent PMN recruitment into the splenic MZ . We further found that complement cooperates with IL-1α-IL-1RI-CXCR2 signaling pathways in recruitment of PMNs to the splenic MZ , which results in elimination of virus-containing MARCO+ macrophages from the spleen . Administration of complement-blocking CR2-Crry or CR2-fH proteins into IL-1α-deficient , but not wild-type , mice prevents PMN accumulation in the splenic MZ and elimination of virus-containing macrophages from the spleen . Our study defines the functional significance of molecular and cellular host defense mechanisms that cooperate in eliminating virus-containing cells in the model of acute disseminated Ad infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biology",
"and",
"life",
"sciences",
"medicine",
"and",
"health",
"sciences"
] | 2014 | IL-1α and Complement Cooperate in Triggering Local Neutrophilic Inflammation in Response to Adenovirus and Eliminating Virus-Containing Cells |
Innate cytokine response provides the first line of defense against influenza virus infection . However , excessive production of cytokines appears to be critical in the pathogenesis of influenza virus . Interferon lambdas ( IFN-λ ) have been shown to be overproduced during influenza virus infection , but the precise pathogenic processes of IFN-λ production have yet to be characterized . In this report , we observed that influenza virus induced robust expression of IFN-λ in alveolar epithelial cells ( A549 ) mainly through a RIG-I-dependent pathway , but IFN-λ-induced phosphorylation of the signal transducer and activator of transcription protein 1 ( STAT1 ) was dramatically inhibited in the infected cells . Remarkably , influenza virus infection induced robust expression of suppressor of cytokine signaling-1 ( SOCS-1 ) , leading to inhibition of STAT1 activation . Interestingly , the virus-induced SOCS-1 expression was cytokine-independent at early stage of infection both in vitro and in vivo . Using transgenic mouse model and distinct approaches altering the expression of SOCS-1 or activation of STAT signaling , we demonstrated that disruption of the SOCS-1 expression or expression of constitutively active STAT1 significantly reduced the production of IFN-λ during influenza virus infection . Furthermore , we revealed that disruption of IFN-λ signaling pathway by increased SOCS-1 protein resulted in the activation of NF-κB and thereby enhanced the IFN-λ expression . Together , these data imply that suppression of IFN-λ signaling by virus-induced SOCS-1 causes an adaptive increase in IFN-λ expression by host to protect cells against the viral infection , as a consequence , leading to excessive production of IFN-λ with impaired antiviral response .
Influenza A virus ( IAV ) , a highly infectious respiratory pathogen , causes worldwide annual epidemics and occasional pandemics . Therefore , IAV has continued to be a top global public health threat . The host cytokine immune response provides the first line of defense against IAV infection . A variety of cell types in the host , including activated alveolar macrophages ( AM ) , lymphocytes , dendritic cells ( DC ) , lung alveolar epithelial cells and endothelial cells within lung tissue , produce cytokines and chemokines following IAV infection , thus playing key roles in innate and adaptive immunity [1]–[3] . However , an aberrant innate response , with early recruitment of inflammatory leukocytes to the lung , was believed to contribute to the morbidity of the 1918 influenza virus infection [4] . Studies have also shown that highly virulent influenza virus infection induces excessive cytokine production ( cytokine storm ) and robust recruitment of leukocytes which are hypothesized to be major contributors to severe disease in humans from influenza virus infection [5] . These data reveal that dysregulation of cytokine signaling of the host during influenza virus infection caused by inappropriate activation of the innate immune response triggers massive pulmonary injury and immune-mediated organ dysfunction . However , the mechanisms underlying the increased induction of innate immune cytokines during influenza virus infection have to date been largely unclear . Innate immune responses triggered by the intracellular detection of viral infections include the production of interferons ( IFNs ) that are classified within the class II cytokine family based on the similarity of their receptors . IFNs consist of three types of cytokines: type I IFNs include IFN-α and IFN-β; type II IFN is IFN-γ and type III IFNs consist of three members in humans , IFN lambda1 ( IFN-λ1 ) , IFN-λ2 , and IFN-λ3 which are also named IL-29 , IL-28A , and IL-28B , respectively , whereas mice only express IFN-λ2 and IFN-λ3 [6] . Virus-infected cells secrete a complex mixture of IFNs that represent a major element of the innate immune response against diverse viral infections [7] . In 2003 , IFN-λs were first discovered as novel antiviral cytokines by two independent groups [8] , [9] . It is now recognized that IFN-λs are virus-induced cytokines with type I IFN-like biological functions , including antiviral activity , but have evolved independently of type I IFNs [10] . Although both type I IFNs and IFN-λ are expressed by a host in response to viral infections , IFN-λs , not type I IFNs , are the predominant IFNs induced by respiratory viruses in nasal epithelial cells and mainly contribute to the first line defense against respiratory virus infection [11] . Type I IFNs were first recognized for their ability to interfere with IAV replication , but IFN-λs have recently been shown to be present at much higher levels than type I IFNs in the lungs of infected mice and play an important role in host defense against IAV infection [2] . However , currently there is limited information available about the biology of IFN-λ . In particular , the mechanisms that regulate the robust expression of IFN-λ during IAV infection are not fully understood . IFN-λs share a common cellular receptor consisting of the cytokine receptor family class II members IL-28R1 and IL-10R2 . The short chain IL-10R2 is ubiquitously expressed and is a receptor component of other type II-related cytokines , whereas the long chain IL-28R1 is unique to IFN-λ and is preferentially expressed on epithelial cells [12] . IFN-λs are induced by most , if not all , classes of viruses as well as some bacterial products [10] . Once secreted , IFN-λs act in an autocrine or paracrine manner by binding the cell-surface receptors . The receptor binding results in a conformational change in the receptor and activation of the receptor-associated Janus tyrosine kinases ( JAKs ) . Activated JAK1 and Tyk2 transphosphorylate the receptor chains that assist in the recruitment of STAT proteins . STAT proteins are then phosphorylated , dimerized , and translocated to the nucleus to initiate transcription of the IFN-stimulated genes ( ISGs ) that mediate the biological effects of IFN-λ . Therefore , IFN-λ-mediated activation of JAK/STAT signaling is required for efficiently triggering the synthesis of antiviral factors . An important mechanism for negative regulation of the JAK/STAT signaling pathway is mediated through members of the SOCS family . Of the eight family members , SOCS-1 has been most extensively studied and is the most potent inhibitor of cytokine-induced signaling [13] . SOCS-1 can directly interact with JAKs by its kinase inhibitory region ( KIR ) , which inhibits JAKs activity . In addition , SOCS-1 can target JAKs to proteasomal degradation through interaction of SOCS box with the Elongin BC complex , which becomes part of an E3 ubiquitin ligase [14]–[16] . When overexpressed in cells , SOCS-1 can inhibit STAT activation induced by multiple cytokine stimulations . Interestingly , several recent studies have revealed that influenza virus has developed mechanisms to subvert host antiviral defense mediated by type I and type II IFNs through inhibition of the JAK/STAT signaling by upregulated SOCS-1 and SOCS-3 proteins [17]–[20] . Consistent with these observations , it has been shown that IFN-λ-induced mRNA expression of the antiviral proteins 2′ , 5′-OAS and Mx1 was abolished by overexpression of SOCS-1 [21] . However , the relationship between suppression of cytokine signaling by SOCS-1 and overproduction of IFN-λ during influenza virus infection remains to be determined . In this study , we examined the effects of influenza virus-provoked negative regulation of cytokine signaling on the IFN-λ production by altering expression of SOCS-1 and activation of STAT signaling . We found that disrupting SOCS-1 expression or constitutive activation of STAT1 significantly inhibits production of IFN-λ in vitro and in vivo . The results reveal that suppression of innate immune cytokine signaling by virus-induced SOCS-1 contributes to formation of IFN-λ storm during influenza virus infection .
To investigate the mechanisms by which the host cells interact with influenza A virus ( IAV ) , we have recently used cDNA microarray to profile the cellular transcriptional response to A/WSN/33 influenza virus ( H1N1 ) infection in A549 human alveolar epithelial cells [22] . Surprisingly , we found that IL-28A , IL-28B and IL-29 , three recently discovered IFN-λ family members , were most significantly up-regulated ( Figure S1A ) . This finding was confirmed by independent experiments measuring the mRNA levels by quantitative real time PCR of IAV infected A549 cells and mouse lungs ( Figure 1A and B ) and RT-PCR ( Figure S1B ) , and evaluating the IL-29 protein level by ELISA ( Figure S1C ) . Treatment of IAV at 56°C for 30 minutes , which prevents viral replication without affecting viral entry into host cells [23] , significantly reduced the virus-induced production of IFN-λs ( Figure 1C ) . To further determine whether production of IFN-λ was affected by viral entry into host cells , IAV was inactivated at 65°C for 30 minutes , which denatures hemagglutinin ( HA ) and prevents host cell attachment [24] . We found that expression of IFN-λ induced by 65°C-inactivated virus recapitulated that of the non-infected control ( Figure 1C ) . These experiments demonstrated that robust expression of IFN-λs was the response to live virus entry into host cells and viral replication . To further determine the inducer of IFN-λs , A549 cells were transfected with either different amounts of total RNA isolated from IAV infected cells ( Figure 1D and Figure S1D ) or genomic RNA directly isolated from the viruses ( Figure S1E ) . The results revealed that both viral genome RNA and viral RNA generated during IAV infection contributed to IFN-λ production . Unlike cellular RNA , influenza viral RNA contains a 5′-triphosphate group which is thought to be the critical trigger for production of type I IFNs through RIG-I-dependent pathway [25]–[27] . Using calf intestine alkaline phosphatase ( CIAP ) to remove the 5′-triphosphate terminus of viral RNA , we tested whether it was involved in IFN-λ induction . Interestingly , treatment with CIAP greatly inhibited expression of IFN-λs ( Figure 1E and Figure S1F ) . To determine whether IAV-induced expression of IFN-λ was completely dependent on RIG-I , A549 cell lines stably expressing shRNAs targeting either RIG-I , TLR3 or MDA5 were generated ( Figure 1F and Figure S1G ) . We observed that silencing RIG-I resulted in a marked decrease in the production of IFN-λ and silencing TLR3 slightly decreased the IFN-λ levels , whereas disruption of MDA5 expression had no overt effects on the IFN-λ production ( Figure 1G–H and Figure S1G–H ) . These data suggest that IFN-λ induced by the IAV RNA was mainly through a pathway involving RIG-I . In normal cells , the strength and duration of cytokine signaling are tightly regulated . However , little is known about why a huge amount of IFN-λ is induced during IAV infection . To address this issue , we sought to investigate whether regulation of IFN-λ-mediated signaling was altered during the viral infection . IFN-λ , like type I IFN , primarily activates the JAK-STAT signal pathway to achieve its antiviral function . Unlike type I IFN receptor , IFN-λ receptor is expressed in a cell-specific fashion [28] . Here we observed that IFN-λ was able to activate JAK-STAT signal pathway in A549 cells ( Figure 2A , B ) . Furthermore , the level of IFN-λ-induced STAT1 phosphorylation was markedly reduced in IAV infected cells , as compared with that in control cells ( Figure 2B–D ) . To substantiate this finding , a time course experiment was performed . We found that phosphorylation of STAT1 in infected cells was dramatically inhibited at later stages of infection ( after 15 h p . i . ) ( Figure 2E ) , while no significant decrease in STAT1 phosphorylation was observed in the cells treated with corresponding culture supernatants ( SN ) from the infected cells ( Figure 2F ) . These data indicate that activation of JAK-STAT signaling by IFN-λ was suppressed in the presence of IAV . Next , we further investigated how IAV infection inhibits IFN-λ-induced STAT1 phosphorylation in A549 cells . Of the eight members of SOCS family , SOCS-1 is the most potent inhibitor of cytokine-induced signaling . In addition , it has recently emerged that SOCS-1 is an important regulator of innate immune response triggered by IAV [18] . Therefore , we hypothesized that SOCS-1 is involved in inhibition of STAT1 phosphorylation during IAV infection . To test this , SOCS-1 mRNA levels in A549 cells during IAV infection were examined by quantitative RT-PCR ( Figure 3A ) . The mRNA level of SOCS-1 was significantly up-regulated at early stages and began to reduce at late stages of infection , but its protein level was consistently increased ( Figure 3B ) . Immunofluorescence study showed that increased expression of SOCS-1 and inhibition of STAT1 phosphorylation occurred specifically in IAV infected cells ( Figure S2A , S2B ) . This implies that there might be a certain relationship between expression of SOCS-1 protein and inhibition of STAT1 phosphorylation . Surprisingly , although SOCS-1 expression in A549 cells was induced by supernatants derived from infected cell culture at later stages ( Figure 3C , D and Figure S2C ) , the SOCS-1 expression induced by IAV infection appeared earlier than that triggered by cell culture supernatants ( Figure 3C , D ) , and than the initial production of IL-29 protein ( Figure S2D ) . The results strongly suggest that during IAV infection , there was a cytokine-independent mechanism to induce SOCS-1 expression . To further verify the functional involvement of SOCS-1 in the suppression of STAT1 activation , SOCS-1 expression in A549 cells was knocked down by shRNA ( Figure 3E ) . In SOCS-1 knockdown A549 cells , but not the control cells , the level of STAT1 phosphorylation was notably increased during the infection ( Figure 3F , G ) , indicating that SOCS-1 was a direct inhibitor of STAT1 phosphorylation during IAV infection . Since JAK-STAT signaling was inhibited by IAV-induced SOCS-1 , we asked whether the activated JAK-STAT pathway by IFN-λ is also disrupted by SOCS-1 . To address this issue , we examined the effect of SOCS-1 protein on activation of STAT1 by IL-29 . As shown in Figure 4A , down-regulation of SOCS-1 enhanced IL-29-induced activation of STAT1 , whereas overexpression of SOCS-1 inhibited IL-29-induced activation of STAT1 in A549 cells , revealing that SOCS-1 negatively regulates IL-29-mediated STAT1 signaling . Moreover , in IAV-infected SOCS-1-ablated A549 cells , STAT1 phosphorylation was markedly elevated by IL-29 stimulation , and this effect was prolonged in both infected and uninfected cells when compared to the control cells ( Figure 4B , C ) . These findings suggest that SOCS-1 inhibits IL-29 signal pathway during IAV infection . Since IAV-induced expression of SOCS-1 appeared earlier than expression of IFN-λ ( see above description ) , it was interesting to investigate whether the induced SOCS-1 influenced IFN-λ production . Surprisingly , the protein level of IL-29 was significantly reduced in the SOCS-1-ablated cells , comparing with that in the control cells infected with IAV ( Figure 4D ) . Furthermore , the mRNA levels of IL-29 and IL-28A/B were also significantly reduced in SOCS-1-ablated A549 cells ( Figure 4E , F ) , while the mRNA levels of ISGs ( MX1 and OAS-2 ) did not change significantly at this time point ( Figure 4E , F ) . The results suggest that the antiviral response is not affected despite less production of IFN-λ in SOCS-1-ablated cells . Because the results presented above revealed that IFN-λ-mediated activation of STAT1 was abrogated during IAV infection , next we asked whether forced activation of STAT1 had any effect on expression of IFN-λs . To test this possibility , we generated A549 cell lines stably expressing either empty vector ( EV ) , STAT1 wild type ( WT ) , or constitutively activated form STAT1-2C ( 2C ) [29] , [30] . The enhancement of STAT1 phosphorylation during IAV infection or stimulated by IFN-λ was confirmed in STAT1-2C-expressing cells ( Figure 5A , B ) . STAT1 phosphorylation was also increased in infected cells overexpressing STAT1-WT as compared to the control cells ( Figure 5B ) . Interestingly , production of IL-29 protein was remarkably decreased in the STAT1-2C-expressing cells as compared to the control after IAV infection ( Figure 5C ) . Consistent with this observation , the mRNA levels of IL-29 and IL-28A/B were significantly reduced in IAV infected STAT1-2C-expressing cells ( Figure 5D , E ) . Furthermore , we tested whether alteration of IFN signaling had any effect on IFN-α and IFN-β production . We found that silencing SOCS-1 or overexpression of STAT1 slightly reduced the type I IFN production during IAV infection ( Figure S3A–C ) . On the other hand , no significant change in the induction of OAS-2 and Mx1 was observed in these cells at late time points post infection ( Figure 5D , E ) . Interestingly , at early time point post infection , activation of STAT1 signaling promoted expression of OAS-2 and Mx1 ( Figure S3D–F ) . In an attempt to provide insights into the mechanism of how inhibition of cytokine signaling causes excessive expression of IFN-λ during IAV infection , we evaluated the pathway governing IFN-λ expression . We found that level of viral RNA , the inducer of IFN-λ expression was unchanged by silencing SOCS-1 expression or forcing STAT1 activation ( Figure S3G , S3H ) . Furthermore , forced activation of cytokine signaling did not alter expression of Pattern-Recognition Receptors ( PRRs ) including RIG-I and TLR3 ( Figure S3H ) . Expression of TLR-7/8 was also examined but they were undetectable in A549 cells . Since alteration of cytokine signaling did not affect the levels of PRRs and viral RNA and given that IRF3 is a known regulator of IFN expression at the early stage of infection , we determined whether there was a functional link between IFN-λ signaling and activation of nuclear factor of κB ( NF-κB ) , a key transcriptional factor downstream of RIG-I pathway [31] . To this end , cells were infected with IAV using increasing MOI . Interestingly , experiment using luciferase reporter gene revealed a positive correlation between increased NF-κB activity and increased expression of SOCS-1 and IFN-λ in infected cells ( Figure 6A , B ) . By contrast , STAT1 phosphorylation and IκB protein levels were consistently reduced ( Figure 6C ) . To further confirm this finding , we employed the A549 cell lines stably expressing SOCS-1 shRNA or active form of STAT1 . We observed that in infected cells , depletion of SOCS-1 increased IκB protein level ( Figure 6D ) and significantly decreased NF-κB activation ( Figure 6E ) . Similarly , forced activation of STAT1 inhibited the degradation of IκB ( Figure 6F ) , and as a result , activation of NF-κB was significantly suppressed in the infected cells ( Figure 6G ) . In contrast , low IκB level and high level of NF-κB activation were detected in SOCS-1-overexpressing cells after IAV infection even using low MOI ( Figure S4A , S4B ) . Consistent with these observations , immunofluorescence microscopy study showed that nuclear translocation of NF-κB p65 was significantly abrogated in SOCS-1-ablated or STAT1-activated cells infected with IAV ( Figure 6H , I and Figure S4C , S4D ) . Together , these results suggest that disruption of cytokine signaling pathway results in robust activation of NF-κB , which causes excessive production of IFN-λ during IAV infection . To confirm the correlation of type III IFN expression with the activation of STAT-1 and NF-κB signaling , a time course study was performed in more detail in infected cell culture ( Figure 7A ) . The results indicated that disruption of IFN-λ signal by SOCS-1 increased their expression likely through activating NF-κB in IAV infected cells . Next , we sought to determine the expression levels of SOCS-1 in mouse lung at different stages of IAV infection . We found that MLD50 of the WSN virus was approximately 3×103 pfu under our conditions , consistent with the previous observation [32] . Therefore , mice were inoculated intranasally with 1×105 pfu of the virus ( about 33 MLD50 ) as previously described [33] , [34] . As shown in Figure 7B , expression of SOCS-1 protein was consistently increased during 3 days of infection . As a consequence , STAT1 phosphorylation was inhibited ( Figure 7B ) . Moreover , activity of NF-κB was elevated during the infection as indicated by the gradually diminished IκBα levels , suggesting that the expression kinetics of IFN-λ correlated with NF-κB activation . Of interest , expression of SOCS-1 was earlier and faster than that of IFN-λ ( Figure S5A , B ) , suggesting that SOCS-1 expression is cytokine-independent at least at the early stage of infection and SOCS-1 might regulate IFN-λ expression beyond negative feedback regulation to respond the cytokines in vivo . This finding is consistent with our in vitro results presented above ( Figure 3C , D ) . To further address the relationship between the expression of SOCS-1 and the induction of IFN-λ , membrane-permeable peptides of SOCS-1-KIR and pJAK2 were used to mimic SOCS-1 overexpression and counteract SOCS-1 function , respectively . The functions of these peptides in IFN-λ response were confirmed in vitro , as the phosphorylation of STAT1 stimulated by IL-29 was dramatically inhibited in the presence of SOCS-1-KIR but not the control peptide SOCS-1-KIR2A ( Figure 7C ) , and the inhibitory effect of SOCS-1-KIR on STAT1 phosphorylation was markedly diminished when SOCS-1-KIR was added together with pJAK2 peptide ( Figure 7F ) . When mice were treated with these peptides and then inoculated with IAV , IFN-λ level was significantly increased in mice treated by SOCS-1-KIR ( Figure 7D and Figure S5C ) . By contrast , the expression of IFN-λ in mice treated with pJAK2 peptide was significantly reduced as compared to control group ( Figure 7G and Figure S5D ) . Furthermore , low levels of STAT1 phosphorylation and IκBα protein were found in SOCS-1-KIR treated mice after IAV infection ( Figure 7E ) , whereas high levels of STAT1 phosphorylation and IκBα protein were present in pJAK2 treated group ( Figure 7H ) . In addition , our experiments showed that treatment with SOCS-1-KIR increased mouse body weight loss , whereas pJAK2 treatment reduced the body weight loss during IAV infection ( Figure S5E–F ) . Together , these data suggest that JAK-STAT signaling pathway is disrupted by increased SOCS-1 in infected mice , which results in an increase in IFN-λ expression likely through activating NF-κB . To further define the role of SOCS-1 in IFN-λ production induced by IAV , we wished to establish a more physiological model system for analysis of SOCS-1 involvement in this process . For this , SOCS-1-knockdown transgenic mice ( TG ) were generated as previously described ( Figure 8A–D ) [35] , [36] . The transgenic founders with high interference efficiency were selected ( Figure 8C , D ) . We found that the level of STAT1 phosphorylation was greatly increased in TG compared to wild type ( WT ) mice after IAV infection ( Figure 8E ) . In contrast , the activity of NF-κB was reduced as indicated by increased IκBα level . Consistent with this , expression of IFN-λ was significantly decreased in IAV infected TG mice ( Figure 8F , G ) . Furthermore , by haematoxylin and eosin ( HE ) staining , we found that on Day 3 p . i . , the lung in mice showed obvious inflammation , but the inflammation in the lung of SOCS-1 knockdown TG mice was minor compared to WT control ( Figure S6A ) . Less body weight loss was observed and Less viral load was detected in the lung of TG mice than that in WT group ( Figure 8H and Figure S6B–C ) , suggesting that although IFN-λ expression is reduced in SOCS-1 knockdown TG mice , the innate antiviral immune response is enhanced .
The clearance of IAV during infection depends on the activation of effective innate and adaptive immune responses . Cytokines activate innate immune responses and initiate the development of adaptive , virus-specific immune responses [37] , [38] . Thus , cytokines play critical roles in defense against the virus infection . Various types of cells in host secrete cytokines and chemokines following IAV infection . Among these cell types , epithelial cells are thought to be one of the most important cytokine-producing cells during IAV infection , and believed to be vital for the virus-induced cytokine storm [39] . We have previously profiled the cellular transcriptional response to IAV infection in human type II alveolar epithelial cell line A549 and found that this type of cell expresses many different cytokines and chemokines after the virus infection [22] . In this study , we show that IAV infection induces excessive expression of IFN-λ that is mainly dependent on RIG-I signaling and partially on TLR3 signaling , indicating that they are involved in the innate antiviral response to the infection . This observation is consistent with previous studies showing that IFN-λ are the predominant IFNs induced by respiratory viruses and have a wide range of antiviral functions in response to respiratory viruses [11] , hepatitis C virus [40] , rotavirus [12] , herpes simplex virus [41] and influenza virus [42] . IFN-λ receptor complex is composed of the ubiquitously expressed short chain IL-10R2 and the long chain IL-28R1 expressed preferentially on epithelial cells [12] . IFN-λs bind the receptors to activate the JAK-STAT signaling pathway which initiates transcription of the ISGs . Thus , IFN-λs , like other types of IFNs , play important roles in the control of IAV propagation in epithelial cells [42] , [43] . Since IFNs are important in a variety of cellular processes , their production and response are delicately regulated by multiple mechanisms . Viruses have evolved different ways to counteract these mechanisms , leading to dysregulation of IFN expression and function , and then successfully evaded the host antiviral response . IAV also exerts its effects through some mechanisms . For example , the viral non-structural protein 1 ( NS1 ) has been shown to inhibit type I IFN response and block IFN-β production . On the other hand , it has also been shown that the capacity of NS1 to confer resistance to host immune response by decreasing sensitivity to particular cytokines causes their overproduction [39] . It remains an ongoing task to determine whether overproduction of IFN-λs is regulated by the NS1 . Recently , it has been shown that IAV induces expression of SOCS-1 and SOCS-3 to negatively regulate JAK-STAT pathway and thereby down-regulates the innate immune response including abrogation of the type I IFN signaling [17] . In the present study , we found that SOCS-1 was greatly induced before the abundant secretion of cytokines in both IAV-infected A549 cells and IAV-infected mice , strongly indicating that during IAV infection , there is a cytokine-independent mechanism to provoke SOCS-1 expression at least at the early stage . Similarly , it has been reported that early induction of SOCS-3 transcription is IFN-β-independent [17] . However , Julien and coworkers have observed that up-regulation of SOCS-1 and SOCS-3 in IAV-infected cells is IFNAR1-dependent [18] , which does not contradict with our observation , because we found that the culture supernatants at the later stages of infection indeed stimulated SOCS-1 expression . Therefore , we conclude that IAV might induce cytokines-independent SOCS-1 expression through other mechanisms , and at least this is true at the early stage of IAV infection . Our experiments demonstrated that IAV-provoked STAT1 phosphorylation at the early stage of infection was inhibited by the virus-induced SOCS-1 . Furthermore , we provided evidence that JAK-STAT signaling activated by IFN-λ was also inhibited by SOCS-1 . It has been previously shown that IAV abrogates the innate immune response mediated by type I IFNs and IFN-γ by disruption of the JAK-STAT signaling pathway [17] , [20] . However , little is known about how suppression of cytokine signaling by SOCS proteins affects the production of IFNs during IAV infection . Interestingly , here we found that the IFN-λ levels were significantly decreased in IAV infected SOCS-1-depleted A549 cells and transgenic mice as compared to infected controls . Importantly , forced activation of STAT1 also significantly inhibits the production of IFN-λ in vitro and in vivo . Despite decreased expression of IFN-λ , the antiviral response was not impaired in SOCS-1-depleted cell and animal . These results suggest that suppression of IFN-λ signaling by SOCS-1 results in their excessive production during IAV infection . Our hypothesis is that suppression of cytokine signaling by virus-induced SOCS-1 leads to an adaptive increase in IFN-λ production by host to protect cells against viral infection . However , increased IFN-λ further induces the expression of SOCS-1 at late stage of infection , which in turn , inhibits the activation of JAK-STAT signaling . Finally , this vicious cycle results in the excessive production of IFN-λ with an impaired antiviral activity due to increased SOCS-1 protein during IAV infection . Although we observed that forced activation of IFN signal also slightly decreased the levels of type I IFNs , whether this hypothesis applies to other cytokine storm provoked by highly virulent influenza virus infection is unclear . In addition , we found that after IAV infection , the SOCS-1 knockdown transgenic mice did not display a remarkable phenotype as compared to wild type mice . However , it is possible that SOCS-1-mediated upregulation of IFN-λ levels has a more prominent role in pathogenesis of highly pathogenic strains of IAV that elicit hypercytokinemia and lethal phenotypes . These remain to be further determined . Our study has also begun to address the mechanism by which inhibition of cytokine signaling causes the excessive expression of IFN-λ during IAV infection . We presume that the repression of STAT1 might activate other transcriptional factors to elevate cytokine levels . Previous studies have shown that aryl hydrocarbon receptor couples with STAT1 to regulate lipopolysaccharide-induced inflammatory responses [44] . It has also been revealed that progressive dysregulation of NF-κB and STAT1 leads to pro-angiogenic production of CXC chemokines [45] . It is thought that NF-κB and STAT1 might have a crosstalk [46]–[48] . Although it is unclear how the phosphorylation of STAT1 can be associated with NF-κB activation [49] , our data showed that IAV inhibited STAT1 activation but promoted the degradation of IκBα , and thus the activity of NF-κB was enhanced both in vitro and in vivo . Moreover , our results revealed that IκBα was degraded and thereby the activity of NF-κB was increased when SOCS-1 was up-regulated by IAV . In fact , this finding is consistent not only with the enhancement of NF-κB activity in SOCS-1 overexpressed-keratinocytes after stimulation by the poly- ( I∶C ) , but also with the increased NF-κB activation in SOCS-1-transfected cells [18] , [50] . Together , these data suggest that suppression of cytokine signaling by SOCS-1 may influence the NF-κB activation . Further research is required to address how inhibition of JAK-STAT signaling is involved in regulation of NF-κB activation .
The mouse experimental design and protocols used in this study were approved by “the regulation of the Institute of Microbiology , Chinese Academy of Sciences of Research Ethics Committee” ( Permit Number: PZIMCAS2012001 ) . All mouse experimental procedures were performed in accordance with the Regulations for the Administration of Affairs Concerning Experimental Animals approved by the State Council of People's Republic of China . Influenza virus strain A/WSN/33 ( H1N1 ) was prepared as previously described [22] , [51] . For infection , cells were washed with phosphate-buffered saline ( PBS ) and infected with the multiplicity of infection ( MOI ) as indicated in the figure legends . After adsorption with α-MEM medium containing 2 µg/ml TPCK ( L-1-tosylamido-2-phenylethyl chloromethyl ketone ) -treated trypsin , 100 U/ml penicillin , and 100 µg/ml streptomycin for 45 minutes at 37°C , the supernatant was aspirated and cells were cultured with the α-MEM medium for indicated time . To inactivate the viruses , equal amounts of viruses were incubated at 56°C or 65°C for 30 minutes as described previously [24] . Plasmids pRC-CMV-STAT1-WT and pRC-CMV-STAT1-2C in which N658 and A656 of STAT1 were substituted by cysteine residues were kindly provided by Dr . David A . Frank ( Dana-Farber Cancer Institute , Boston , MA ) . The cDNA coding STAT1-WT or STAT1-2C was subcloned into the Not I/Sal I sites of retroviral vector pMSCV-IRES-GFP ( pMIG ) to generate pMIG-STAT1-WT and pMIG-STAT1-2C . The vector pMIG-SOCS-1 was previously described [13] . NF-κB-luciferase reporter named pNF-κB-Luc and Renilla luciferase reporter named pRL-TK were gifts from Dr . Shijuan Gao ( Institute of Microbiology , Chinese Academy of Sciences ) . For luciferase assay , cells were co-transfected with pNF-κB-Luc , pRL-TK and indicated plasmids , and luciferase activity was measured using the dual-luciferase reporter assay system according to the manufacturer's instruction ( Promega , U . S . ) . The following antibodies were used in this study: anti-STAT1 ( E23 ) , anti-phospho-STAT1 ( Tyr701 ) , anti-RIG-I , anti-NF-κB p65 ( Santa Cruz Biotechnology , Santa Cruz , CA ) ; and anti-β-actin ( Abcam ) . All other antibodies were obtained as previously described [13] , [51] . Peptides of SOCS-1-KIR ( ( 53 ) DTHFRTFRSHSDYRRI ) , SOCS-1-KIR2A ( ( 53 ) DTHFATFASHSDYRRI ) and pJAK2 ( ( 1001 ) LPQDKEYYKVKEP ) were synthesized by ChinaPeptides ( Shanghai , China ) . All peptides were synthesized with an attached lipophilic group , palmitic acid , to facilitate entry into cells as previously described [16] , [52] . Peptides were purified by preparative RP-HPLC , and were characterized by LC-MS and HPLC analysis . Recombinant human IL-28A and IL-29 were purchased from PeproTech ( Rocky Hill , NJ ) . Cells were incubated with the recombinant IL-29 ( 50 ng/ml ) for 45 minutes for stimulation , unless otherwise indicated . Supernatant culture medium from the A549 cells infected with IAV strain A/WSN/33 ( H1N1 ) was also used as a source of virus-induced cytokines for cell stimulation . To quantify IL-29 production by host cells , supernatant culture medium from virus infected cells was harvested and examined by enzyme-linked immunosorbent assay ( ELISA ) using the ready-SET-Go of human IL-29 analysis kit ( eBioscience , San Diego , CA ) according to manufacturer's instruction . Total RNA was prepared from A549 cells infected with the IAV for 8 hours ( viral RNA ) or from uninfected cells ( cellular RNA ) using Trizol ( TIANGEN BIOTECH BEIJING CO . , LTD . ) according to manufacturer's instructions . The calf intestine alkaline phosphatase ( CIAP ) ( TaKaRa ) was used to dephosphorylate viral 5′-triphosphate RNA as previously described [17] . A549 cells were transfected with the isolated RNA using Lipofectamine 2000 ( Invitrogen ) . Supernatant medium from transfected cells was harvested and examined by ELISA for IL-29 production . The transfected cells were lysed and examined by real-time PCR for expression of indicated genes . For Western blotting analysis , cell lysates were separated by SDS-polyacrylamide gel electrophoresis , transferred onto a nitrocellulose membrane , and probed with indicated antibodies as described previously [53] . To detect nuclear translocation of NF-κB p65 , immunofluorescence was performed as described previously [22] . Images were acquired using a confocal microscope ( Model LSCMFV500 ) and a 60× oil immersion objective lens ( both from Olympus Optical , Japan ) with an NA of 1 . 40 . Short hairpin RNA ( shRNA ) -based knockdown cell lines were generated by infection of A549 with lentiviruses expressing specific shRNA in pSIH-H1-GFP vector as described previously [22] . The sequences used in the shRNAs targeting specific genes were as follows: human SOCS-1 shRNA#1 5′-GCATCCGCGTGCACTTTCA-3′ [54] and shRNA#2 5′-CTACCTGAGCTCCTTCCCCTT-3′ [55] , mouse SOCS-1 shRNA#1 5′-GGACGCCTGCGGCTTCTAT-3′ and shRNA#2 5′-CTACCTGAGTTCCTTCCCCTT-3′ [56] , human RIG-I shRNA#1 5′-TGCAATCTTGTCATCCTTTAT-3′ and shRNA#2 5′-AAATTCATCAGAGATAGTCAA-3′ [57] , and human TLR3 shRNA#1 5′-GGTATAGCCAGCTAACTAG-3′ and shRNA#2 5′-ACTTAAATGTGGTTGGTAA-3′ and human MDA5 shRNA#1 5′-CCAACAAAGAAGCAGTGTATA-3′ [58] and luciferase ( Luc ) control shRNA 5′-CTTACGCTGAGTACTTCGA-3′ . A549 cell lines stably expressing STAT1-WT , STAT1-2C , SOCS-1 or empty vector ( EV ) were generated by infecting the cells with retroviruses encoding these genes in pMIG vector as previously described [13] . Female BALB/c mice ( 5–6 weeks old , 18–20 g ) were provided by Vital River Laboratory Animal Center ( Beijing , China ) . To determine the 50% mouse lethal dose ( MLD50 ) of the virus , six groups of five mice were inoculated intranasally with 10-fold serial dilutions of virus . MLD50 titres were calculated by the method of Reed and Muench [59] . For infection , mice were inoculated intranasally with 1×105 plaque-forming units ( pfu ) of the A/WSN/33 virus . For the peptide treatment , 2 days before viral infection , mice were pre-administrated intraperitoneally ( i . p . ) once a day with peptide using 5 mg/kg body weight . On the indicated day of post-infection ( p . i . ) , the mice were then euthanized and their lungs were removed for further analysis by Western blotting and RT-PCR . SOCS-1-knockdown transgenic mice were generated by the microinjection method as previously described [35] , [36] . Briefly , shRNA-expressing vector targeting mouse SOCS-1 was linearized by Sca I . The transgenic mice were generated by the microinjection and genotyped by PCR using specific primers of up-stream: 5′-AAATCCTGGTTGCTGTCTCTTTATGG-3′ and down-stream: 5′-GGAAGGTCCGCTGGATTGA-3′ . A 350 bp fragment of the shRNA cassette was amplified , which represented integration of the transgenic DNA . Transgenic mice were analyzed by Western blotting using the anti-SOCS-1 antibody . The transgenic founders with high interference efficiency were selected and maintained on a BALB/c genetic background . Comparison between groups was made using Student's t-test . Data represent the mean ± SD . Differences were considered statistically significant with P<0 . 05 . | Influenza virus infection triggers innate immune responses . However , aberrant host immune responses such as excessive production of cytokines contribute to the pathogenesis of influenza virus . Type III interferons ( IFN-λ ) constitute the major innate immune response to influenza virus infection , but the precise pathogenic processes of IFN-λ production and mechanistic underpinnings are not well understood . In this study , we report that influenza virus induces robust IFN-λ expression mainly through a RIG-I-dependent pathway , but signaling activated by IFN-λ was dramatically inhibited by virus-induced SOCS-1 . Importantly , we found that disruption of the SOCS-1 expression or forced activation of STAT1 significantly reduced the expression of IFN-λ in vitro and in vivo , suggesting that suppression of IFN-λ signaling by SOCS-1 results in their excessive production during influenza virus infection . Furthermore , our experiments revealed that disruption of IFN-λ signaling pathway resulted in the activation of NF-κB that governs the IFN-λ expression . Together these findings , we propose that impaired antiviral response of IFN-λ due to the inhibitory effect of SOCS-1 causes an adaptive increase in IFN-λ expression by host to protect cells against the viral infection . This is a novel mechanism that may be critical in the pathogenesis of the influenza virus strains that induce hypercytokinemia . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2014 | Suppression of Interferon Lambda Signaling by SOCS-1 Results in Their Excessive Production during Influenza Virus Infection |
p53 protects us from cancer by transcriptionally regulating tumor suppressive programs designed to either prevent the development or clonal expansion of malignant cells . How p53 selects target genes in the genome in a context- and tissue-specific manner remains largely obscure . There is growing evidence that the ability of p53 to bind DNA in a cooperative manner prominently influences target gene selection with activation of the apoptosis program being completely dependent on DNA binding cooperativity . Here , we used ChIP-seq to comprehensively profile the cistrome of p53 mutants with reduced or increased cooperativity . The analysis highlighted a particular relevance of cooperativity for extending the p53 cistrome to non-canonical binding sequences characterized by deletions , spacer insertions and base mismatches . Furthermore , it revealed a striking functional separation of the cistrome on the basis of cooperativity; with low cooperativity genes being significantly enriched for cell cycle and high cooperativity genes for apoptotic functions . Importantly , expression of high but not low cooperativity genes was correlated with superior survival in breast cancer patients . Interestingly , in contrast to most p53-activated genes , p53-repressed genes did not commonly contain p53 binding elements . Nevertheless , both the degree of gene activation and repression were cooperativity-dependent , suggesting that p53-mediated gene repression is largely indirect and mediated by cooperativity-dependently transactivated gene products such as CDKN1A , E2F7 and non-coding RNAs . Since both activation of apoptosis genes with non-canonical response elements and repression of pro-survival genes are crucial for p53's apoptotic activity , the cistrome analysis comprehensively explains why p53-induced apoptosis , but not cell cycle arrest , strongly depends on the intermolecular cooperation of p53 molecules as a possible safeguard mechanism protecting from accidental cell killing .
The prominence of the p53 gene in tumor suppression is emphasized by its unsurpassed mutation rate in cancer cells [1] . As a master regulatory transcription factor for anti-proliferative programs , p53 can decide cell fate in response to a broad range of stress stimuli , including DNA damage and oncogene activation [1] , [2] , . p53 prevents the accumulation of precancerous cells by activating genes involved in cell cycle arrest ( e . g . p21/CDKN1A , GADD45A , SFN , E2F7 ) and apoptosis ( e . g . BAX , PMAIP1/NOXA , PUMA ) or repressing cell proliferation genes [5] . While gene activation is well-studied , the mechanism of p53-dependent target gene repression is still poorly understood and both direct and indirect models are discussed [5] , [6] . On the one hand , p53 prevents genes from becoming activated by directly binding to promoters or distal enhancer elements - thereby competing with other activating transcription factors and components of the basic transcriptional machinery - or by recruiting histone-modifying enzymes with repressive functions such as mSin3A [5] . On the other hand , p53 indirectly represses proliferation genes by upregulation of several coding ( p21/CDKN1A , E2F7 ) and non-coding RNAs ( miR-34 family , lincRNA-p21 ) [7] , [8] , [9] , [10] , [11] , [12] . Sequence specific DNA binding of p53 requires a DNA motif that consists of two decameric half-sites ( RRRCWWGYYY; R = A/G , W = A/T , Y = C/T ) separated by an optional spacer of additional base pairs to form a full-site [13] . Previous in vitro studies demonstrated that the central CWWG defines the torsional flexibility of the DNA and thus influences p53's binding affinity [14] . While a CATG sequence is flexible and therefore bound with high affinity , the other possible CWWG sequences are not [15] . In fact , it has been suggested that the inflexible CWWG sequences and spacer containing sites require a higher binding energy and therefore represent low affinity p53 binding sites [14] , [15] , [16] , [17] . Interestingly , high affinity p53 motifs are specifically enriched among pro-arrest genes , whereas the promoters of pro-death targets predominantly contain low affinity sites [13] , [16] , [18] . Despite these biophysical differences between p53 binding sequences , it remains unclear at present how p53 molecularly distinguishes between distinct target genes to bind and activate a selected set . Structurally , p53 proteins assemble into an asymmetric tetramer that can be described as a dimer of symmetric dimers . Tetramerization is mediated via the C-terminal oligomerization domains and further stabilized through interactions between neighboring DNA binding domains [19] , [20] . In detail , oppositely charged amino acids ( Glu180 , Arg181 ) in the H1 helices of the DNA binding domains form an inter-molecular double salt bridge that enables adjacent p53 molecules to interact and cooperate when binding to DNA – a property known as DNA binding cooperativity ( Fig . 1A ) [21] , [22] , [23] , [24] . Of note , cooperativity has been shown to be required for p53-induced apoptosis but not cell cycle arrest [24] , [25] . Furthermore , somatic p53 mutations resulting in reduced cooperativity are found in cancer patients , germline cooperativity mutations segregate with cancer susceptibility in Li-Fraumeni syndrome families , and cooperativity mutant mice are highly cancer prone , indicating that DNA binding cooperativity is essential for proper tumor suppression [24] , [25] . The aim of this study was to comprehensively characterize the impact of DNA binding cooperativity on all p53 binding sites in the genome ( the p53 cistrome ) by combined analysis of global DNA binding ( ChIP-seq ) and expression data . We demonstrate that high DNA binding cooperativity is crucial for the binding and transactivation of low affinity binding sites in pro-apoptotic genes with non-canonical and spacer-containing p53 motifs and also for p53-mediated repression of mitotic and pro-survival genes . Since both transactivation of genes with non-canonical response elements and p53-mediated gene repression are essential for p53-induced apoptosis , these data comprehensively explain why p53 molecules need to cooperate for cell killing as the basis for efficient tumor suppression .
To explore the role of DNA binding cooperativity for the genome-wide binding pattern of p53 , we comprehensively mapped the binding sites of p53 proteins in different cooperativity states by deep sequencing of immunoprecipitated chromatin ( ChIP-seq ) . The p53 cooperativity mutation “EE” ( p53R181E ) causes four negatively charged glutamic acid residues to cluster at the H1 helix interaction interface which strongly destabilizes the intermolecular interactions and reduces DNA binding cooperativity [24] . Likewise , the mutation “RR” ( p53E180R ) brings four positively charged arginine residues together resulting in a similar destabilization and low degree of cooperativity . Importantly , combined expression of EE and RR ( EE/RR ) results in mixed tetramers in which one negatively and one positively charged H1 helix interact , resulting in a DNA binding cooperativity that slightly exceeds that of the wild-type ( wt ) ( Fig . 1A ) . Combined expression of EE and RR therefore rescues the cooperativity defect of the EE and RR homotetramers . Following transfection in p53-negative Saos-2 cells all p53 variants were expressed at equal levels comparable to endogenous p53 in U2OS cells treated with the MDM2 inhibitor nutlin-3a ( Fig . 1B ) [26] . ChIP sequencing resulted for each sample in more than 30 M reads that were mapped to the genome . p53 binding peaks were called applying a stringent false discovery rate ( FDR ) of 10−5 . Moreover , only peaks with a minimum number of 50 reads and 2-fold change versus GFP and input were considered as binding sites . These criteria ensured the identification of only reliably p53 bound regions . In detail , 88 peaks were determined as EE binding sites and additional 1579 sites were bound by RR , which together represent 1667 low cooperativity peaks ( Fig . 1C ) . 3145 additional sites occupied by wild-type p53 together with 375 sites bound by EE/RR only , formed the 3520 high cooperativity peaks . Thus , the number of p53 binding sites rises with increasing DNA binding cooperativity as illustrated in the peak density plot ( Fig . 1D ) . When binding sites were ranked according to decreasing EE/RR binding strength as a measure of binding affinity , EE and RR sites clustered at the top ( see heatmap in Fig . 1D ) , indicating that only the high affinity p53 binding sites that were strongly bound by EE/RR or wild-type p53 were also bound by low cooperativity p53 . Accordingly , the binding strength of wild-type p53 to high affinity ( low cooperativity ) sites was significantly stronger than to low affinity ( high cooperativity ) sites that were only occupied by wild-type p53 and/or EE/RR ( Fig . 1E ) . These correlations between cooperativity and binding strength were also evident on the single gene level ( Fig . 1F ) and confirmed in independent validation experiments ( Fig . 1G ) . In summary , the genome-wide ChIP analysis revealed that DNA binding cooperativity extends the number of p53 sites by enabling recruitment to low affinity sites . To explore the location of p53 binding sites in the genome , we divided the genome into the regions: gene body , promoter , distal and intergenic ( Fig . 2A ) . p53 binding sites of both , low and high affinity regions were preferentially located directly at the promoter or within the gene body ( above 70% in each group ) and only rarely at further distance indicating that low and high cooperativity sites are distributed similarly across the genome ( Fig . 2B ) . As a previous ChIP-on-Chip analysis restricted to promoter regions of the genome suggested that the DNA binding cooperativity of p53 influences the sequence preference of p53 [24] , we further characterized the p53 sequence motif in high and low cooperativity p53 binding sites of our genome-wide ChIP-seq dataset . A de novo motif search within all groups of p53 peaks - independent of the level of cooperativity - revealed a p53 motif with significant similarity to the consensus p53 motif ( JASPAR database ) ( Fig . 2C ) . p53 motifs identified in the group of low cooperativity sites showed high uniformity ( E-value 5×10−177 ) while motifs in high cooperativity sites were more divers ( E-value 2 . 4×10−76 ) with the most variability in the subgroup of EE/RR-only bound peaks ( E-value 3 . 5×10−37 ) . In fact , the EE/RR-only motif was more similar to a p53 half-site than to a full-site . Furthermore , a search for p53 motifs on the basis of the wild-type p53 consensus from the present ChIP-seq analysis ( Fig . 2D ) revealed a strong preference for an A in the core CWWG sequences of each half-site that became less obvious with increasing cooperativity ( Fig . 2E ) . To directly compare the quality of p53 motifs in the different cooperativity groups , we scored every single motif instance on the basis of similarity to the wild-type p53 consensus motif using two independent algorithms ( Fig . 2F ) . Approximately 50% of the low cooperativity p53 binding sites matched perfectly to the consensus in contrast to less than 20% of the high cooperativity sites ( Fig . 2F , right ) . In parallel , the mean motif score as determined by the p53MH algorithm [27] decreased with increasing cooperativity ( Fig . 2C , bottom ) . Moreover , whereas spacer sequences were absent in about half of the motif instances in the low cooperativity peaks , 70 to 80% of the motifs identified in high cooperativity peaks contained spacers of variable length ( Fig . 2C , bottom ) . Together , the cistrome analysis suggests that p53 with low DNA binding cooperativity only binds to full-site p53 DNA motifs with high similarity to the consensus binding sequence . In contrast , motifs occupied by highly cooperative p53 only , show reduced similarity to the p53 consensus motif and comprise not only full but also half-sites separated by spacers of variable length . Thus , p53 requires high DNA binding cooperativity for binding to non-canonical p53 motifs . The genes closest to the p53 binding sites were functionally annotated using a combination of different algorithms . Expectedly , Ingenuity Pathway Analysis identified “p53” and “p53 signaling” as the most significant transcriptional regulator and canonical pathway , respectively , in both cooperativity groups ( Fig . 3 ) which validates the two gene lists as significantly enriched for bona fide p53 target genes . Interestingly , the top biological function for low cooperativity target genes was cell cycle progression , in contrast to apoptosis for high cooperativity genes . The cooperativity-dependent difference in biological function was further confirmed by overlap analysis with gene sets in the Molecular Signature Database ( MSigDB , Fig . 3 ) and functional annotation with Gene Ontology terms ( Fig . 3 ) . Both p53 binding site groups showed the strongest overlap with an experimentally defined set of p53-induced genes [28] . Importantly , genes with high affinity p53 sites were again annotated with cell proliferation , stress and immune responses and included well-characterized cell cycle arrest genes such as CDKN1A and BTG2 [29] , [30] . In contrast , genes with low affinity p53 peaks were associated with cell death , and apoptosis in particular , including critical pro-apoptotic genes . Cooperativity-dependent regulation of pro-apoptotic genes was validated for BAX and PUMA/BBC3 , two well-established apoptotic target genes of p53 ( Fig . 3B , C ) [5] . Thus , the requirement for DNA binding cooperativity , which determines the affinity towards different p53 motifs , functionally separates the p53 cistrome into cell cycle regulation and apoptosis . The cistrome of wild-type p53 has been previously characterized in a number of different cell types under various p53-activating conditions [31] , [32] , [33] . To explore the impact of p53 DNA binding cooperativity in a broader context , we compared our data obtained from Saos-2 osteosarcoma cells to p53 ChIP-seq data obtained in the breast cancer cell line MCF7 treated with 5-fluorouracil ( 5FU ) or MDM2 inhibitors ( nutlin-3a , RITA ) [32] and the osteosarcoma cell line U2OS treated with actinomycin D or etoposide [31] ( Fig . 4 ) . In both MCF7 and U2OS cells the p53 cistrome was strongly influenced by the type of p53-activating stimulus and only subsets of all binding peaks were bound in a treatment-independent manner: 3550 MCF7 common and 1611 U2OS common peaks , . Furthermore , the comparison of MCF7 and U2OS cells revealed a pronounced cell type-specificity of the p53 cistrome so that only 1003 common peaks were bound by p53 in both cells types . 719 ( 71 . 7% ) of these common peaks were also present in Saos-2 cells , strongly supporting the hypothesis , previously raised by Nikulenkov et al . [32] , that there is a ‘default set’ of p53 binding sites in the genome that is bound largely independent of treatment and cell type . Many of the other p53 peaks that we identified in Saos-2 cells were also present in MCF7 or U2OS cells , but often only in one cell-type or following a specific treatment , as indicated in Fig . 4 by the percentage of overlap . When analyzing the relative proportion of low and high cooperativity peaks within the overlap , we found that the common ‘default set’ of binding peaks was mostly comprised of low cooperativity peaks , while the overlap with cell type- or treatment-specific peak sets showed a higher percentage of high cooperativity peaks . These data suggest that the ‘default program’ of p53 activation , that possibly functions as a first-line defense to genomic damage , does not require DNA binding cooperativity , while a fine-tuned p53 response , that integrates context-specific cues in a cell type- and stress-dependent manner , strongly relies on DNA binding cooperativity . To investigate the role of DNA binding cooperativity for gene regulation , the p53 cooperativity mutants were analyzed by microarray-based expression profiling in combination with ChIP-seq . 351 genes that were bound by at least one of the p53 versions were found to be differentially regulated by more than 2-fold ( Fig . 5A , Supplemental Table S1 ) . As shown above for the complete p53 cistrome ( Fig . 1E ) , DNA binding cooperativity determined the binding strength also in the regulated part of the cistrome , i . e . the subset of p53-bound and -regulated genes ( Fig . 5A , C ) . Interestingly , the vast majority of these genes ( 97% ) was p53-induced and not repressed . Although the EE mutant was identified on a small number of genes by ChIP-seq , EE was no potent mediator of gene regulation . For all other p53 proteins the transactivation of individual genes directly correlated with binding strength and the average degree of regulation rose with increasing cooperativity ( Fig . 5A , B ) . Gene regulation by the low cooperativity mutant RR was therefore confined to the subset of genes with high affinity binding sites whereas high cooperativity p53 showed additional regulation of low affinity targets ( Fig . 5A ) . The correlation of DNA binding cooperativity with gene regulation was confirmed in validation experiments by RTqPCR for several genes ( Fig . 5D ) , some of which were previously validated to be bound in a cooperativity-dependent manner ( Fig . 1G ) . Interestingly , RAD54L2 , which recruits all p53 cooperativity mutants to a similar extent , showed nevertheless cooperativity-dependent induction suggesting that the role of cooperativity extends beyond regulation of DNA binding and might affect transactivation by additional mechanisms . Furthermore , although TP53I3 ( PIG3 ) displays cooperativity-dependent recruitment to its binding site , it is a rare example of a target gene that was transactivated independently of cooperativity , indicating that low-level binding of p53 maximally activates some genes already . It is tempting to speculate that this exception is due to the peculiar binding of p53 to a polymorphic pentanucleotide microsatellite in the TP53I3 promoter [34] . We conclude from these data that cooperativity not only increases the binding site spectrum of p53 but also gene regulation with respect to gene number and activation level . We next explored potential sequence , positional and functional differences between low and high cooperativity binding sites in the regulated subset of the cistrome ( Fig . 5E ) . De novo motif search discovered in both cooperativity groups a motif significantly resembling the p53 consensus binding motif . However , the motif was less perfect for the high cooperativity peaks and resembled more a half-site than a full-site . In line with this , more than 50% of the motifs in the low cooperativity group were spacer-free in contrast to only 26% in the high cooperativity group . Different from the genome-wide analysis , the genomic location of low and high cooperativity binding sites across the p53-regulated genes varied substantially . 79% of the low cooperativity sites were located within the promoter and only 14% within the gene body , in comparison to 56% and 30% of the high cooperativity sites , respectively . Importantly , functional annotation again revealed a separation of the bound and regulated genes into distinct gene ontology categories with low cooperativity target genes being annotated with cell cycle regulation and high cooperativity genes with cell death . Together , the DNA binding cooperativity of p53 not only determines the number of genomic binding sites but also the number of regulated genes , the vast majority of which are p53-induced instead of repressed . Moreover , not only DNA binding strength but also the level of transactivation correlates directly with the degree of cooperativity indicating that cooperativity enhances p53's impact on the cistrome and transcriptome . Most importantly , p53-regulated genes with low and high cooperativity binding sites differ significantly in their biological function . Transcriptional activation of the apoptosis program requires a higher degree of intermolecular cooperation likely as a safeguard against accidental elimination of cells . It was previously shown by gene expression profiling that p53 mutant and wild-type breast cancer samples are molecularly distinct and that p53-dependent transcriptional signatures not only predict p53 status but also disease-specific survival [35] . The correlation of superior survival with upregulated expression of p53-induced genes was validated in multiple datasets from independent patient cohorts [36] . Considering the role of DNA binding cooperativity for the regulation of functionally distinct classes of p53 target genes , we explored whether expression of low and high cooperativity genes affects patient survival to a similar extent . Using published microarray-based expression data from breast cancer patients [35] , [37] we employed a previously described gene set enrichment analysis ( GSEA ) approach to assess whether the gene expression profile of a patient is enriched in low and/or high cooperativity p53 target genes [38] . Kaplan-Meier curves showed that upregulated expression of high cooperativity target genes was significantly associated with superior survival ( Fig . 6A ) . Surprisingly , no such correlation was observed for low cooperativity genes ( Fig . 6B ) . These data suggest that not all p53 target genes are equally potent in tumor suppression and that only high cooperativity genes are able to prolong the survival of breast cancer patients . In contrast to half of the p53-induced genes ( 476/995 ) , less than 10% of the repressed genes ( 13/221 ) contained a p53 binding peak in the vicinity ( Fig . 7A , Supplemental Table S2 ) . In two repressed genes a p53 binding peak mapped to a distal enhancer element as defined by H3K4 mono- and dimethylation , H3K27 acetylation and DNase I hypersensitiviy ( Fig . 7B ) suggesting that p53 can mediate repression through interfering with distal enhancer activity as previously shown for mouse embryonic stem cells [39] , [40] , [41] . However , most of the downregulated genes did not contain a p53 binding site . Surprisingly , the level of downregulation nevertheless correlated with DNA binding cooperativity ( Fig . 7C ) . While only three genes ( KLF6 , MYC , UTP15 ) were identified to be repressed by the low cooperativity mutant RR , wild-type p53 and EE/RR robustly repressed ( mean 2-fold downregulation ) multiple genes associated with mitotic progression ( e . g . AURKA , AURKB , CDC20 , CCNB1/2 ) , survival ( BIRC5 ) and developmental regulation ( TGFβ and WNT signaling ) ( Fig . 7D , Supp . Table S2 ) . We conclude that p53-mediated gene repression displays an even higher dependence on cooperativity than gene activation . The lack of p53 binding peaks in the vicinity of most repressed genes suggested that p53-dependent repression is largely indirect . As the level of downregulation was shown to be dependent on the level of cooperativity , we predicted that downregulation is mediated by p53 target genes which are induced in a cooperativity-dependent manner . Several possible candidates that have previously been implicated in p53-mediated repression such as lincRNA-p21 , miR-34a , CDKN1A and E2F7 were all induced by p53 [7] , [8] , [9] , [10] , [12] , but only E2F7 showed a cooperativity-dependent expression pattern on the mRNA level ( Fig . 7E ) . Although CDKN1A was transactivated in a cooperativity-independent manner , its gene product p21 showed a clear cooperativity-dependent induction on the protein level ( Fig . 7E , F ) . To interrogate the role of these proteins for p53-mediated gene repression , we examined the effect of CDKN1A or E2F7 depletion ( Fig . 7F ) . The knock-down of CDKN1A did not prevent repression of MYC , E2F8 or GJB2 , but had a slight de-repressive effect on AURKA and BIRC5 and resulted in complete de-repression of CDC20 ( Fig . 7G ) . Upon depletion of E2F7 some genes were unaffected ( MYC , AURKA , BIRC5 , CDC20 ) whereas repression of E2F8 and GJB2 was strongly reduced ( Fig . 7G ) . Additional bioinformatic analysis with Ingenuity Pathway Analysis and GeneXplain [42] in terms of p53-downregulated genes revealed a significant enrichment of transcription factor binding sites ( Sp1 , SMAD3 , NF-Y ) and common upstream regulators ( YY1 , FOXM1 ) as well as additional miRNAs ( miR-34 , miR-145 , miR-200 ) several of which have been previously implicated in transcriptional repression by p53 [43] . Since p53 engagement of a repressive effector network comprising cell cycle inhibitors , transcriptional repressors , miRNAs and long non-coding RNAs is largely cooperativity-dependent , DNA binding cooperativity is therefore - despite the striking underrepresentation of p53-repressed genes in the p53 cistrome - nevertheless a major determinant of both gene activation and repression .
Cooperative DNA binding by p53 is known to be essential for p53-mediated cell death and cooperativity mutations in cancer patients suggest a role for tumor suppression [24] , [44] . This is further supported by a selective apoptosis defect and cancer susceptibility of cooperativity mutant mice [25] . Here we used p53 H1 helix mutants in genome-wide DNA binding and expression analyses to comprehensively profile the role of DNA binding cooperativity for p53's function . In line with previous data showing that DNA binding of p53R181E ( EE ) is hardly detectable [22] , [24] , we identified only a very small number of 88 EE binding sites in the genome compared to 4812 binding sites for wild-type p53 ( Fig . 1C ) . Although the cooperativity-reducing p53E180R ( RR ) mutation resulted in a similar DNA binding defect as the EE mutation when studied in vitro using recombinant proteins purified from E . coli [22] , the DNA binding defect of RR expressed in mammalian cells was weaker and resulted in a cistrome of 1667 binding sites comprising mostly perfect consensus-like full-site motifs in genes enriched for cell cycle regulators . Importantly , while both EE and RR homotetramers have a more or less reduced cooperativity , they complement each other efficiently to yield EE/RR heterotetramers with a cooperativity higher than wild-type p53 . This increases the efficiency of cooperative DNA binding and enables binding to a larger spectrum of 5188 sites enriched for non-canonical , spacer-containing p53 motifs in , for example , pro-apoptotic genes ( Fig . 2 and 3 ) . Together , these data prove that DNA binding cooperativity is a crucial modulator of p53's genome-wide binding pattern ( cistrome ) with important functional relevance for cell fate determination . Previous ChIP-seq studies of activated wild-type p53 have identified approximately 1800 to 2900 significant binding sites [31] , [32] , which is comparable to the number of 1667 peaks bound by low cooperativity p53 . Direct comparison of ChIP-seq data from different cells treated with multiple p53-activating drugs further revealed that the default set of p53 binding sites common to most cell types and independent of the type of activating stimulus largely comprises low cooperativity sites ( Fig . 4 ) . In contrast , binding sites that were bound by p53 in a cell-type and stress-specific manner were enriched in high cooperativity sites ( Fig . 4 ) , suggesting that fine-tuning the p53 response in a context-specific manner relies on DNA binding cooperativity . On the DNA side , the interaction of p53 with a certain DNA motif is largely influenced by the central CWWG sequence even though the WW dinucleotide is not directly contacted by p53 [15] , [19] , [20] . As proper binding of the p53 tetramer to DNA requires bending of the DNA , different affinities of CWWG sequences can be explained by differences in bending flexibility [14] , [20] , [45] . As CATG is the most flexible CWWG sequence , intermolecular cooperation of p53 monomers is likely dispensable , while efficient binding to the more rigid non-CATG may require higher bending forces that depend on energetic stabilization provided by strong H1 helix interactions [24] , [44] . Consistent with high affinity binding of p53 to CATG , we identified a specific enrichment of central CAWG sequences among the high affinity sites that were bound irrespective of cooperativity ( Fig . 2E ) . In contrast , in line with lower affinity binding of p53 to CAAG , CTTG or CTAG , these non-CATG sequences showed a stronger dependence on cooperativity ( Fig . 2E ) . Together , H1 helix interactions allow p53 molecules to cooperate to provide sufficient energy required for bending and binding a larger variety of sequences in the genome . Importantly , there is growing evidence that such non-canonical , low affinity binding sites contribute substantially to p53's function [18] . First of all , considerable transactivation was observed at non-canonical half-sites ( single decamers ) and three-quarter-sites , some of which were originally classified as biologically relevant response elements ( REs ) in , for example , the pro-apoptotic target genes PIDD and APAF1 [17] . Moreover , REs in many other functionally important pro-apoptotic genes show on average less similarity to the p53 consensus sequence and a lower degree of evolutionary conservation associated with higher sequence diversity than most prototypic cell cycle target genes such as CDKN1A [13] , [16] , [46] . Another example is the VEGFR1 gene promoter , which contains a single nucleotide polymorphism that generates a non-canonical , functional p53 half-site thereby integrating the VEGF system into the p53 transcriptional network [47] . In fact , it is discussed that weak p53 REs have a selective advantage compared with high-affinity p53 binding sites as they could allow better fine-tuning of responses through the regulation of p53 protein levels , specific post-translational modifications or cofactors that modulate DNA binding affinity [18] , [44] , [48] . Crosstalk between p53 and the estrogen receptor in regulating VEGFR1 provides a prominent example for the functional dependence on cooperation of p53 with other transcription factors for maximal activation of such non-canonical response elements [49] . Cooperativity therefore dramatically expands the p53 transcriptional network allowing the engagement of target genes that - likely as a safeguard - require a higher degree of stress or damage for activation . The integrated analysis of ChIP-seq with expression profiling data revealed that less than 10% of the p53-bound genes were regulated by p53 ( Fig . 5 ) . This is in line with other studies [31] , [33] , [50] , [51] , and indicates that p53 binding to DNA is often not sufficient to induce transcription . Secondary stimuli or co-factors are needed , possibly in a stress or cell type specific manner , to induce a permissive chromatin state as previously suggested for single p53 target gene promoters [52] . While binding sites of low and high cooperativity p53 showed a similar distribution across the genome and were in 70–80% located in the promoter region or gene body ( Fig . 2A , B ) , functional binding events that resulted in expression changes were distributed differently . While binding sites regulated by low cooperativity p53 were mainly enriched in the promoter region of genes , binding sites regulated by high cooperativity p53 were also frequently observed in the gene body ( Fig . 5E ) . As binding of p53 to the regulatory promoter is more likely to have a direct effect on transcription than binding to a site further downstream of the transcriptional start site , this finding is consistent with the hypothesis that low affinity , non-canonical binding sites primarily function in cooperation with other transcription factors to fine-tune gene expression in response to context- or tissue-specific stimuli [18] . Interestingly , our analysis on the role of DNA binding cooperativity for patient survival showed a remarkable difference between low and high cooperativity genes . While upregulated expression of high cooperativity target genes correlated with a good clinical outcome , expression of low cooperativity target genes was surprisingly not correlated with distinct patient survival ( Fig . 6 ) . This indicates that low and high cooperativity genes are clinically not equivalent . Although low cooperativity genes comprise the default program of target genes activated in most cell types in a stimulus-independent manner , only the activation of high cooperativity genes is able to prolong patient survival . Together these data strongly emphasize the clinical relevance of DNA binding cooperativity for the anti-cancer activity of p53 . Intriguingly , a number of studies from both breast cancer patients and mice have found a wild-type p53 status to be associated with an inferior clinical response to chemotherapy compared to tumors with mutant p53 [53] , [54] , [55] , [56] . For example , in MMTV-Wnt1 driven mouse mammary tumors p53 wild-type tumor cells can evade an apoptotic chemotherapy response by undergoing arrest , followed by secretion of senescence-associated cytokines that can stimulate proliferation and relapse [53] . Given the functional separation of the p53 cistrome into high cooperativity genes with proapoptotic function and low cooperativity genes involved in cell cycle arrest , it is tempting to speculate that the expression ratio of high versus low cooperativity genes might determine the clinical response to chemotherapy in p53 wild-type tumors . While activation of high cooperativity genes is expected to prolong the long-term survival of the patient by supporting the apoptotic chemotherapy response , activation of low cooperativity genes leading to senescence might even be counter-productive and promote relapse . In fact , there was a trend - although not statistically significant - that in patients without upregulation of high cooperativity genes expression of low cooperativity genes was associated with an inferior survival ( data not shown ) . Although it still remains to be investigated whether cooperativity has a similar impact on patient survival in other tumor entities , it is intriguing that the DNA binding cooperativity of p53 is not only crucial for preventing tumor development [25] but also appears to have a clinical impact on the survival of cancer patients under therapy . Overall 221 genes , significantly enriched for mitotic and developmental regulators , were robustly repressed in response to p53 activation ( Fig . 7 ) . This is consistent with numerous studies that have established repression of cell cycle regulatory genes as a function of p53 [5] . For example , p53 robustly downregulates the MYC oncogene for induction of cell cycle arrest [57] . Surprisingly , in contrast to half of all activated genes , only 13 of 221 repressed genes showed p53 binding in the ChIP-seq experiment ( Fig . 7A ) . Although some studies have reported direct binding of p53 to non-canonical p53 response elements in repressed genes , convincing genome-wide data supporting direct p53 binding as a general mechanism of repression is missing [5] , [6] . Of note , interference of p53 with distal enhancer elements has been described to mediate repression of stem cell specific genes in murine embryonic stem cells [39] . We can confirm p53 binding to distal enhancers in two cases ( ADRB1 , NUFIP1 ) , but overall the mechanism does not seem to play a prominent role in our cell model suggesting cell type specificity ( Fig . 7B ) . Interestingly , despite the absence of p53 binding events at repressed genes , the degree of repression was nevertheless dependent on cooperativity ( Fig . 7 ) . Importantly , p53-mediated repression of anti-apoptotic genes and oncogenes is crucial for p53-induced apoptosis [12] , [58] . As cooperativity-reducing mutations result in apoptosis deficiency [24] and impair both the transactivation of important pro-apoptotic genes and the repression of a large set of genes , it is conceivable that cooperativity-dependent repression contributes to the pro-apoptotic function of p53 . As p53 is not directly binding to most of the downregulated genes ( Fig . 7A ) , an indirect mechanism involving p53-mediated transactivation of genes with repressor functions might be underlying repression mechanistically . In support of this , mice carrying a mutation in the p53 transactivation domain were reported as strongly impaired not only for transactivation but also repression [59] . Furthermore , a number of p53-activated genes including CDKN1A and E2F7 have been implicated in repressing a variety of p53-regulated genes [7] , [8] , [9] . Consistently , we identified many E2F target genes as repressed by p53 and confirmed the requirement of E2F7 and CDKN1A for the repression of mutually exclusive sets of genes ( Fig . 7F , G ) . As both , E2F7 and p21 , are induced in a cooperativity-dependent manner ( Fig . 7F ) , their regulation could contribute to the cooperativity-dependent repression of at least a subset of p53-downregulated genes . Furthermore , p53 has been implicated as a master regulator of miRNAs expression and processing [43] . The by far best studied group of p53-activated miRNAs is the miR-34 family that targets many mitotic genes contributing to senescence and apoptosis induction [10] , [60] , [61] . Other miRNAs induced by p53 such as miR-145 or the miR-200 family have been implicated as inhibitors of MYC and important developmental genes , respectively [43] . Our bioinformatics analysis of the cooperativity-dependently repressed genes revealed several miRNAs as potential upstream regulators , amongst others miR-34 , miR-145 and miR-200 . In addition , p53 was shown to repress genes indirectly by upregulating the large intergenic non-coding RNA lincRNA-p21 , which is believed to interact with chromatin modifying complexes to silence target genes [12] . We observe induction of lincRNA-p21 in our study but do not see a major impact of cooperativity on lincRNA-p21 expression , excluding this lincRNA as a cause of cooperativity-dependent gene repression . In principle , cooperativity-dependent gene repression in the absence of direct p53 binding to the repressed target promoters could alternatively indicate that cooperativity mutations affect other aspects of p53 function apart from DNA binding . While cooperativity mutations - different from many hot-spot mutations - do not affect the overall folding of the DNA binding domain [22] , it is known that amino acids E180 and R181 are engaged in p53 interactions with ASPP family proteins that stimulate p53-transactivation of pro-apoptotic target genes [62] , [63] . Furthermore , it has been described that association of p53 with promoter-specific cofactors like Sp1 , SMAD3 , NF-Y and YY1 results in gene repression [5] , [6] . We applied both a sequence-based promoter analysis and a search for common upstream regulators of the p53-repressed genes and identified a significant enrichment of all these factors . Whether the interaction of Sp1 , SMAD3 , NF-Y and YY1 with p53 is dependent on cooperativity has so far not been explored . A future analysis of the interactome of p53 cooperativity mutants might therefore reveal additional insight into the effect of cooperativity mutations on gene repression . In summary , our combined genome-wide analysis of DNA binding and gene expression using a set of p53 mutants with reduced and increased cooperativity reveals DNA binding cooperativity as a major modulator of the p53 cistrome . In particular the use of high cooperativity p53 enabled the compilation of a comprehensive set of p53 binding sites including many non-canonical response elements that have previously not been profiled . Interestingly , cooperativity is revealed to be not only important for p53 binding to non-perfect response elements but also for p53-mediated gene repression . Since both transactivation of non-canonical response element and p53-mediated repression are crucial for p53's pro-apoptotic activity , this strengthens the concept that the requirement for intermolecular p53 cooperation provides a novel safeguard mechanism protecting against the accidental activation of apoptosis as the most final , irreversible cell fate decision possible .
Saos-2 and U2OS cells were cultured in Dulbecco's modified Eagle's medium ( Life Technologies ) supplemented with 10% fetal bovine serum ( PAA ) and penicillin/streptomycin ( Life Technologies ) using standard conditions and procedures . siRNAs were purchased from Dharmacon and transfected with Lipofectamine RNAiMAX ( Life Technologies ) according to the manufacturer's protocol . Generation and use of recombinant adenoviruses for wild-type p53 and p53R181E and p53E180R have been described [24] . Saos-2 cells were infected with adenovirus encoding GFP ( as a control ) or GFP together with wild-type or mutant p53 . Cells were fixed 18 hours after infection in fresh 1% paraformaldehyde ( PFA ) for 10 min at room temperature ( RT ) . Unreacted PFA was quenched by adding glycine to a final concentration of 125 mM for 5 min at RT . Cells were washed twice with ice-cold PBS and scraped from the dishes in ice-cold phosphate buffered saline supplemented with proteinase inhibitor ( Complete , Roche ) . Cells were pelleted ( 700×g for 5 min at 4°C ) and lysed at a concentration of 2×107 cells/ml in SDS Lysis Buffer ( 1% SDS , 10 mM EDTA , 50 mM Tris pH 8 . 1 ) supplemented with proteinase inhibitor . Cells were sonicated on ice in 1 ml aliquots 5 times at 30% power for 10 sec followed by a 50 sec pause with a SONOPLUS sonifier with sonotrode MS72 ( Bandelin electronics , Germany ) . Agarose gel electrophoresis confirmed shearing of crosslinked DNA into a smear in the range of 200–800 bp . After centrifugation at 10 , 000×g for 10 min at RT supernatants were diluted 1∶10 with Dilution Buffer ( 0 . 01% SDS , 1 . 1% Triton X-100 , 1 . 2 mM EDTA , 16 . 7 mM Tris-HCl pH 8 . 1 , 167 mM NaCl ) and after 1 h of pre-clearing 1% input was removed from each sample and proteins were precipitated with p53 DO-1 antibody over night at 4°C . Mock-chromatin was immunoprecipitated from cells infected with GFP only . Complexes were bound to Protein G magnetic beads ( Fast Flow , GE healthcare ) for 2 h at 4°C and washed once with Low Salt Immune Complex Wash Buffer ( 0 . 1% SDS , 1% Triton X-100 , 2 mM EDTA , 20 mM Tris-HCl pH 8 . 1 , 150 mM NaCl ) , once with High Salt Immune Complex Wash Buffer ( 0 . 1% SDS , 1% Triton X-100 , 2 mM EDTA , 20 mM Tris-HCl pH 8 . 1 , 500 mM NaCl ) , once with LiCl Immune Complex Wash Buffer ( 0 . 25 M LiCl , 1% IGEPAL-CA630 , 1% deoxycholic acid ( sodium salt ) , 1 mM EDTA , 10 mM Tris-HCl pH 8 . 1 ) , and twice with TE ( 10 mM Tris-HCl , 1 mM EDTA , pH 8 . 0 ) for about 5 min at 4°C . Complexes were eluted with Elution buffer ( 1% SDS , 0 . 1 M NaHCO3 ) for 20 min at 800 rpm at RT . Crosslinks were reversed at 65°C in 200 mM NaCl overnight followed by RNase A ( 37°C , 30 min ) and Proteinase K digestion ( 45°C , 2 h ) . DNA was purified using the PCR Purification Kit ( Qiagen ) and DNA concentration was measured with PicoGreen ( PicoGreen dsDNA Quantitation reagent , Molecular Probes ) . The enrichment was verified by qPCR for known binding sites . For primer sequences see Supplemental Table S3 . For each sample , a single library was sequenced once on an Illumina GA IIx ( ChIP-Seq Sample Preparation Kit , Cluster Generation Kit v2 , 36-Cycle Sequencing Kit v3 ) and twice on an Illumina HiSeq2000 ( TruSeq SR Cluster Kit v3 - cBot - HS , TruSeq SBS Kit v3 - HS - 50 cycles ) system . RNA isolation and cDNA synthesis were performed using the RNeasy Mini Kit ( Qiagen ) and SuperScript VILO cDNA Synthesis Kit ( Life Technologies ) . miRNA isolation was performed using the mirVana miRNA Isolation Kit ( Life Technologies ) followed by reverse transcription using the TaqMan miRNA Reverse Transcription Kit ( Life Technologies ) . Gene expression was quantified by RTqPCR using SYBR Green Jumpstart Taq ReadyMix ( Sigma ) or TaqMan MicroRNA Assays ( U6 snRNA , miR-34a; Life Technologies ) on a LightCycler 480 ( Roche ) . Expression data were normalized to GAPDH or U6 snRNA and the mock sample using the ΔΔCt method . For primer sequences see Supplemental Table S3 . Cells were lysed in 50 mM Tris-HCl pH 8 . 0 , 150 mM NaCl , 5 mM EDTA pH 8 . 0 , 2% Nonidet P-40 and the total protein concentration was quantified by Bradford assay . Samples were separated by SDS–PAGE and transferred onto nitrocellulose membranes ( GE healthcare ) . After blocking with 10% non-fat dry milk , membranes were probed with antibodies specific for p53 DO-1 ( gift from B . Vojtesek ) , CDKN1A ( C-19 , Santa Cruz ) , E2F7 ( H-300 , Santa Cruz ) , or β-actin ( AC15 , Abcam ) . Enhanced chemiluminescence ( Thermo Scientific ) or fluorescence ( Odyssey Infrared Imaging System , LI-COR ) was used for detection . ChIP-seq data have been deposited in the EBI ArrayExpress archive ( http://www . ebi . ac . uk/arrayexpress/ ) and are accessible through the accession numbers E-MTAB-1394 ( Username: Reviewer_E-MTAB-1394; Password: 0vmiovxP ) Microarray data sets have been deposited in the EBI ArrayExpress archive ( http://www . ebi . ac . uk/arrayexpress/ ) and are accessible through the accession number E-MTAB-1403 ( Username: Reviewer_E-MTAB-1403; Password: srkaaska ) | The tumor suppressor gene p53 counteracts tumor growth by activating genes that prevent cell proliferation or induce cell death . How p53 selects genes in the genome to direct cell fate specifically into one or the other direction remains unclear . We show that the ability of p53 molecules to interact and thereby cooperate , influences which genes in the genome p53 is regulating . In the absence of cooperation , p53 only binds and regulates a limited ‘default’ set of genes that is proficient to stop cell proliferation but insufficient to induce cell death . Cooperation increases p53's DNA binding and enables context-dependent activation of apoptosis genes and repression of pro-survival genes which together triggers cell death . As the concerted effort of p53 molecules is needed , the threshold for cell killing is raised possibly to protect us from accidental cell loss . Thus , by shaping the genomic binding pattern , p53 cooperation fine-tunes the gene activity pattern to steer cell fate into the most appropriate , context-dependent direction . The genome-wide binding patterns of cooperating and non-cooperating p53 proteins generated in this study provide a comprehensive list of p53 binding sites as a resource for the scientific community to further explore mechanisms of tumor suppression by p53 . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"oncology",
"medicine",
"tumor",
"physiology",
"gene",
"expression",
"biology",
"basic",
"cancer",
"research",
"molecular",
"cell",
"biology",
"dna",
"transcription"
] | 2013 | Characterization of the p53 Cistrome – DNA Binding Cooperativity Dissects p53's Tumor Suppressor Functions |
Fv1 is the prototypic restriction factor that protects against infection by the murine leukemia virus ( MLV ) . It was first identified in cells that were derived from laboratory mice and was found to be homologous to the gag gene of an endogenous retrovirus ( ERV ) . To understand the evolution of the host restriction gene from its retroviral origins , Fv1s from wild mice were isolated and characterized . Most of these possess intact open reading frames but not all restricted N- , B- , NR-or NB-tropic MLVs , suggesting that other viruses could have played a role in the selection of the gene . The Fv1s from Mus spretus and Mus caroli were found to restrict equine infectious anemia virus ( EIAV ) and feline foamy virus ( FFV ) respectively , indicating that Fv1 could have a broader target range than previously thought , including activity against lentiviruses and spumaviruses . Analyses of the Fv1 sequences revealed a number of residues in the C-terminal region that had evolved under positive selection . Four of these selected residues were found to be involved in the novel restriction by mapping studies . These results strengthen the similarities between the two capsid binding restriction factors , Fv1 and TRIM5α , which support the hypothesis that Fv1 defended mice against waves of retroviral infection possibly including non-MLVs as well as MLVs .
Viruses co-evolve with their hosts , upon which they are completely dependent for replication . As the host acquires strategies to restrict virus infection the invaders develop counter measures to evade restriction . The ensuing genetic conflict can play out over an extensive timeframe [1] , [2] , [3] , [4] . Due to the unique replication strategy employed by retroviruses where integration of viral genetic information into the host genome occurs [5] , the conflict between virus and host can take an interesting twist . When integration occurs in germ or embryonic cells , the virus can become an endogenous retrovirus ( ERV ) and inherited through the germ line [6] , [7] . As a result , viral gene products can be conscripted to serve as defensive forces against further viral infection [8] . The murine retrovirus restriction gene , Fv1 , provides perhaps the prototypic example of one such gene [9] . Fv1 restriction was first described in the early 1970s [10] , [11] as an activity protecting mice against infection with murine leukemia virus ( MLV ) . Two semi-dominant alleles were identified , Fv1n and Fv1b , that provide protection against B-tropic and N-tropic MLVs , respectively [12] , [13] . The crucial difference between N-tropic and B-tropic MLV maps within the viral gag gene to a single codon encoding amino acid 110 of the mature capsid ( CA ) protein [14] indicating that CA represents the target for the restriction factor . MLVs insensitive to Fv1 , called NB-tropic , carry further changes in CA [15] , [16] . The mode of action of the Fv1 protein is not fully understood but indirect evidence suggests that it binds to CA on the cores of incoming virions shortly after virus entry into the cell without inhibiting viral reverse transcription [17] but somehow preventing entry of newly synthesized viral DNA into the nucleus [9] . Based on sequence similarity , Fv1 appears to be derived from the gag gene of an ancient ERV called MERV-L ( murine endogenous retrovirus with a leucine tRNA primer binding site ) though it appears only distantly related to MLV [18] . Amino acid 110 of CA also determines sensitivity of MLV to another retrovirus restriction factor , TRIM5α [19] , best known for its ability to restrict HIV-1 [20] . While there is no similarity between Fv1 and TRIM5α at the primary sequence level , both molecules share a similar domain organization [9] . The N-terminal domains both contain an essential coiled coil motif involved in multimerization while the respective C-terminal domains are required for specific virus binding [21] , [22] . Indeed , the C-terminal domain of Fv1 can be replaced with CypA , a molecule that binds HIV-1 CA , resulting in a factor that restricts HIV-1 [23] . TRIM5 has been isolated from a number of mammals including a variety of primates , rabbits and cows [20] , [24] , [25] , [26] , [27] . These have been shown to restrict a range of retroviruses from different genera . In particular , TRIM5 from the cotton top tamarin can restrict gammaretroviruses , lentiviruses , spumaviruses and betaretroviruses [24] , [28] , [29] . Comparison of the target sequences show little identity and although the gammaretroviral , betaretroviral and lentiviral CA molecules show a similar tertiary structure [30] the spumavirus target is folded very differently [31] . Residues in the C-terminal B30 . 2 domain of TRIM5α that determine viral recognition , and thus restriction specificity , are under strong positive selection [32] , [33] and are thought to evolve under pressure imposed by retroviral infection [3] , [34] , [35] . However in no case have the viruses involved been identified unambiguously [36] , [37] , [38] . In contrast , changes in Fv1 and the acquisition of its antiviral activity are less well defined . Based on its distribution in different subgenera of Mus , it appears that the Fv1 gene was inserted around 4–7 million years ago [39] , [40] . However this finding is somewhat paradoxical because the only known target for Fv1 , MLV , probably arose considerably more recently as judged by the distribution of its endogenous forms [41] , [42] . What then drove the spread and survival of the Fv1 open reading frame ? Could it be that viruses other than MLV selected for Fv1 ? To address this question we have developed a panel of Fv1 genes from different mice and investigated their anti-viral activity against a variety of retroviruses . These studies reveal an extraordinary degree of plasticity in the Fv1 gene as well as two non-MLV viral targets suggesting that a number of different viruses have moulded its evolution .
To study the evolution of Fv1 , we set out to clone the gene from a variety of species of Mus . Consistent with previous reports [39] , [40] , it proved possible to clone Fv1 from multiple species of the subgenus Mus as well as single examples of the subgenera Mus nannomys and Mus pyromys ( Table 1 ) . However , we failed to amplify Fv1 from Mus coelomys , Apodemus and Rattus despite multiple attempts [43] , suggesting that the insertion leading to Fv1 arose about five million years ago , at the time when the ancestors of Pyromys , Nannomys and Coelomys diverged [44] . Sequencing revealed open reading frames in all cases except Mus mus terricolor ( dunni ) and M . m . cookii ( Figure S1 ) . In M . m terricolor , this was due to a single base pair deletion at position 224 that causes a frameshift and premature stop , while in M . m . cookii , a base pair transition from C to T at position 650 coupled with a 5 base pair deletion causes the formation of a premature stop codon . Interestingly , in three cases , M . m . molossinus , M . m . spretus and M . m . caroli , 2 different sequences were amplified in reproducible fashion . Pairs clustered together in phylogenetic analyses , suggesting the presence of more than one segregating allele in these subspecies of mice . Sequence comparisons reveal that the N-terminal region of Fv1 , which encodes an extended coiled coil region necessary for restriction activity [23] , [45] , is well conserved ( Figure S1 ) . Compared to M . m . caroli , M . m . famulus , Mus nannomys minutoides and Mus pyromus platythrix , all the other Fv1s contained a 3 amino acid insertion near the N-terminus ( Figure S1 ) . This change was not important for restriction activity . By contrast , the C-terminal domain shows significant variation in regions important for Fv1 function . Four variable regions , which we designate VA–D can be distinguished ( Figure 1 ) . The first variable region ( residues 247–276 ) overlaps a sequence called the Major Homology Region ( MHR ) that is present in the CA protein of all retroviruses as well as Fv1 [18] , [46] , [47] and is essential for Fv1 function [48] . Variable regions B–D ( amino acids 345–358 , 375–401 and the extreme C-terminus ) contain the residues we had previously shown to distinguish the predicted products of the n and b alleles of Fv1 . These differences are found at amino acids 358 , 399 and the very C-terminus of the Fv1 protein where an apparent deletion of 1 . 3 kb in genomic DNA resulted in a nineteen amino acid length difference [18]; together they appear responsible for the differences in restriction specificity [48] . The present analysis showed that the more divergent mice contained the residues that are found in Fv1n at positions 358 and 399 but they did not contain the 1 . 3 kb deletion . This suggested that Fv1n arose from the progenitor Fv1 , which was similar in length to Fv1b , through an internal deletion , while Fv1b evolved through the substitution of the residues at positions 358 and 399 . Variable regions A–C appear to arise by point mutation but region D shows more significant changes in nucleotide sequence . The three most distantly-diverged mice , M . n . minutoides , M . m . famulus and M . p . platythrix each appear to have B1 repeat sequences inserted , apparently independently , near the deletion site that gave rise to the Fv1n allele ( Figure 2 ) . They contribute the last few amino acids of Fv1 resulting in C-termini that are rather different from either Fv1n or Fv1b . Other differences in this region arise from short insertions or deletions perhaps resulting from polymerase slippage during DNA replication . Thus the clones , SPR1 and SPR2 , we amplified from M . m . spretus of French and Spanish origins differed by four amino acids; the same difference also seen between Fv1b and CAS2 ( Figure 2 ) . By analogy with other restriction factors it seems likely that the variation seen in Fv1 arose by selection and this would be reflected in changes in restriction specificity . To test this we examined the restriction properties of Fv1 we had cloned from different mice ( Table 1 ) as well as a number of genes synthesized on the basis of published sequences ( Table 2 ) [40] , [49] . A tree based on these sequences is shown in Figure 3; it shows good agreement with the accepted phylogeny of genus Mus [50] . The Fv1 genes were introduced into pLgatewayIYFP and tested for restriction of different MLVs using a two-colour FACS assay [19] , [51] . The results are shown in Table 3 . As previously shown , the Fv1n gene restricted B-MLV but not N- or NB-MLV while Fv1b , when expressed at protein levels seen in transduced cells , restricted N- and NB- MLV and also had a weak activity against B-MLV [51] . The Fv1 gene from M . m . molossinus and M . m . spretus restricted both N- and B-MLV but not NB-MLV , suggesting that they could be similar to Fv1nr [16] . Hence , we also examined three N-MLV CA variants ( D82N , H114R and L117H ) that confer resistance to Fv1nr [16] . Variants N-MLV H114R and N L117H , which we have found to be NR tropic , were also resistant to both the Fv1MOL1 and Fv1SPR1 proteins . In contrast , N-MLV N82D was restricted by Fv1MOL1 but not by Fv1SPR1 , suggesting that Fv1MOL1 is subtly different from Fv1nr . We have cloned the Fv1nr gene from 4 different strains of mice ( 129SvEv , NZB/B1NJ , NZW/LacJ and RF/J ) ; all contained a single nucleotide change compared to Fv1n , causing a serine to phenylalanine substitution at residue 352 . While Fv1MOL1 also encoded phenylalanine at position 352 , Fv1SPR1 possesses a serine at the corresponding position . Taken together , these results suggest that other changes could also be involved in determining the nr-specificity . Perhaps surprisingly , Fv1 from two closely related species , M . m . castaneus and M . m . spicelegus lacked perceptible Fv1 activity against MLV . Other Fv1 genes displayed a variety of restriction phenotypes . Fv1 from two asian members of the Mus subgenus , M . m . caroli and M . m . cervicolor , lacked Fv1 activity directed against MLV as did the two members of the Pyromys subgenus that we tested , M . p . platythrix and M . p . saxicolor ( Table 3 ) . By contrast two members of the Nannomys subgenus , M . n . ninutoides and M . n setulosis were active with the M . n . minutoides clones restricting all six MLVs tested . The M . m . famulus sample , whose position in Mus phylogenetic trees is relatively poorly defined , showed weak activity against N , B , and NB-tropic MLVs . Thus more than half of the Fv1 genes with intact open reading frames did not seem to have any activity against MLV , the target that defines the Fv1 gene , even though they were expressed at similar levels to restricting genes in transduced cells ( Figure S2 ) . Further , the extent and specificity of restriction of different MLVs varies significantly . Clearly , the properties of the restriction gene have changed since the gene first became part of the mouse germ line but whether MLV alone was responsible for selecting such changes remained an open question . Prompted by the example of TRIM5α that can restrict multiple genera of retrovirus [24] , [29] , we decided to investigate the hypothesis that non-MLV retroviruses might play a role in shaping the evolution of Fv1 , by testing a number of different retroviral vectors for restriction by Fv1 from wild mice . These included other gammaretroviruses like Gibbon Ape Leukemia Virus ( GALV ) , Feline leukemia Virus ( FeLV ) and Porcine Endogenous Retrovirus-A ( PERV-A ) , lentiviruses such as HIV-1 , HIV-2 , SIVmac , Equine Infectious Anemia Virus ( EIAV ) and Feline immunodeficiency Virus ( FIV ) , as well as foamy viruses including Prototypic Foamy Virus ( PFV ) , Simian Foamy Virus ( SFV ) and Feline Foamy Virus ( FFV ) . Some of these results are presented in Table 4 . The data show that Fv1 from M . m . caroli , that lacked activity against MLV , restricted FFV strongly and PFV weakly . Moreover , Fv1 from M . m . spretus , which restricted N- and B-MLV , and from M . m . macedonicus , which inhibited N-MLV , were also active against the lentivirus EIAV . By contrast GALV , FeLV , PERV-A , SFV , HIV-1 , SIVmac and FIV were not restricted by any of the Fv1 genes in the panel ( Table 4 and data not shown ) . Formally it remains possible that the novel specificities observed result from over expression . Unfortunately , no cell lines expressing Fv1CAR1 and Fv1SPR1 at endogenous levels are available , precluding a direct test of this idea . However we are not aware of any examples of complete restriction of novel viruses resulting from such a mechanism . To further characterize restriction mediated by Fv1CAR1 and Fv1SPR1 stable cell lines were derived by transducing MDTF cells with retroviral vectors carrying these genes and selecting for G418-resistant single cell clones . These cell lines were used in virus titrations by measuring the percentage of transduced cells by FACS with different amounts of virus ( Figure 4A , B ) . As expected , the titre of EIAV was dramatically reduced in the cell line expressing Fv1SPR1 compared to the untransduced MDTF control ( Figure 4A ) . Similarly titres of FFV and PFV were greatly reduced in MDTF cells expressing Fv1CAR1 compared to untransduced while titres of SFV were unaffected by the presence of the Fv1 gene ( Figure 4B ) . These results confirm the observations made with the 2 colour FACS assay that Fv1 from some wild mice can restrict non-MLV retroviruses . Fv1 is thought to interfere with MLV replication by preventing nuclear import of newly synthesized viral DNA [9] . To test whether this was also true for EIAV and FFV , we examined the fate of viral DNA in restricting cell lines . Testing EIAV replication in Fv1SPR1 cells shows no inhibition of reverse transcription as measured by levels of newly synthesized late DNA products ( Figure 4C ) . However levels of 2-LTR circles , which are thought to form only after nuclear entry [52] , are substantially reduced suggesting a block in nuclear uptake . In Fv1CAR1 cells a reduction in FFV 2-LTR circles , with no change in late RT products , was also observed ( Figure 4D ) , again consistent with a block in nuclear import . However , interpretation of these data is complicated by the fact that the majority of FFV DNA synthesis is thought to occur in the producer cells [53] . Nevertheless , it appears likely that Fv1 is acting to block lentivirus and foamy virus replication at the same stage in the viral life cycle as seen with MLV . To identify the specificity determinants of these novel restriction activities , chimeric Fv1 genes were constructed and tested for restriction . To look at FFV restriction , we made chimeras between Fv1CAR1 , which restricted only FFV , and Fv1n , which restricted B-MLV . Schematic views of the constructs made and the corresponding restriction data are shown in Figure 5A . Replacement of a C-terminal fragment of Fv1n ( from residue 318 ) with the corresponding fragment from Fv1CAR1 generated a chimera ( Fv1nC4 ) capable of restricting FFV . Replacement with a shorter fragment starting from residue 353 ( Fv1nC5 ) was insufficient to confer restriction , suggesting that the determinants of FFV restriction were found between residues 316 and 352 of Fv1 from M . m . caroli . In the reciprocal chimeras , replacing the small C-terminal segment of Fv1CAR1 beginning from residue 352 with that from Fv1n did not result in any loss of activity against FFV . However , when a larger fragment beginning at residue 316 was replaced , activity was lost , confirming the presence of the determinants of FFV restriction within the region of Fv1CAR1 between residues 316 and 352 . Within this region , there are 5 residues that differ between Fv1n and Fv1CAR1 . These were systematically changed to identify the residues involved in specificity determination ( Figure 5B ) . No single change could endow Fv1n with the ability to restrict FFV ( Figure 5B ) . However two single changes at positions 349 and 352 of Fv1CAR1 resulted in loss of FFV restriction . We therefore mutated both these positions in Fv1n to the corresponding amino acids found in Fv1CAR1 . This generated a construct ( Fv1nE349KS352Y ) capable of restricting FFV . Taken together , these results indicate that both lysine 349 and tyrosine 352 in Fv1 from M . m . caroli are crucial for FFV restriction . We had previously shown that MLV recognition maps downstream of this region [48]; it was therefore interesting to see that Fv1nE349KS352Y ( and chimera Fv1Cn5 ) could recognize both B-MLV and FFV in an additive fashion . To examine EIAV restriction by Fv1 from M . m . spretus , a second set of chimeras was made between Fv1SPR1 , which restricts N-MLV , B-MLV and EIAV , and Fv1n , which only restricts B-MLV . Restriction of EIAV was seen with chimeras only when amino acids from positions 191 and 271 were derived from Fv1SPR1 ( Fig . 6A ) suggesting that the determinants of EIAV restriction lay between these residues . Interestingly , the determinants for MLV restriction were slightly different from those of EIAV . Replacing a short segment of C-terminus of Fv1n ( from residue 366 ) with that from Fv1SPR1 in Fv1nS3 was sufficient to confer restriction of N-MLV , suggesting that this region contained determinants of N-MLV restriction . However , a reciprocal change in Fv1SPR1 ( Fv1Sn3 ) did not abolish N-MLV restriction . It was only when a C-terminal segment beginning with residue 191 was replaced from Fv1SPR1 ( Fv1Sn1 ) that the restriction of N-MLV was lost . This suggested that additional requirements for N-MLV restriction were found between residues 191 and 271 of Fv1 from M . m . spretus , perhaps overlapping with those that determined EIAV restriction . There are 5 differences between Fv1n and Fv1SPR1 in the segment between residues 191 and 271 ( Figure 6B ) . To identify the residues involved in restriction , site-directed mutagenesis was employed to change the residues in Fv1n to those present in Fv1SPR1 . Reciprocal mutations were also made in Fv1SPR1 . These mutants were tested for restriction of EIAV , N- and B-MLV . The substitution from arginine to cysteine at position 268 in Fv1n was sufficient to confer the ability to restrict both N-MLV and EIAV . The reciprocal change in Fv1SPR1 resulted in a partial reduction in restriction of all three viruses . These results indicated that residue 268 was the major determinant of EIAV restriction by Fv1SPR1 and had an influence on MLV restriction but that other neighbouring residues were also important . A lysine to glutamine change at residue 270 in Fv1n resulted in low but reproducible restriction of EIAV and N-MLV though the reciprocal change in Fv1SPR1 had little effect . Interestingly , substitution of the residue at position 261 in both Fv1n and Fv1SPR1 seemed to abolish the restriction of B-MLV , indicating that this residue was involved in the interaction with B-MLV . We conclude that residues 261 , 268 and 270 in Fv1 from M . m . spretus are all involved in virus recognition . However , it would appear that recognition of EIAV by Fv1 from M . m . macedonicus has arisen in a different manner as it contains arginine rather than cysteine at position 268 .
In this study of Fv1 evolution we have demonstrated that Fv1 shows substantial sequence variation in its C-terminal half , the region of the protein thought to contain determinants of restriction specificity . In addition we have shown that Fv1 is capable of restricting viruses other than its previously defined targets and identified the sequence variation responsible for these novel targets . We note that some Fv1 alleles do not appear to possess an associated restriction activity; it would be of considerable interest to determine whether they recognize other targets . A previous study had identified six codons , specifying Fv1 amino acids 261 , 265 , 270 , 362 , 299 and 401 , that show evidence for positive selection during the course of Mus evolution [40] . These represent potential sites of interaction between Fv1 and its target viruses . Combining these data with our previous studies of Fv1 specificity [16] , [48] , it seems reasonable to conclude that the four variable regions defined in Figure 1 constitute four domains collectively or individually involved in target selection and binding ( Figure 7 ) . Thus VRA ( amino acids 247–276 ) includes the positively selected residues 261 , 265 and 270 as well as three residues , 261 , 268 and 270 , shown to be important for EIAV restriction by Fv1SPR1 while VRB ( amino acids 345–358 ) has positively selected amino acid 352 , amino acids 349 and 352 important for FFV recognition by Fv1CAR1 as well as residues 352 and 358 important for NR- and N- versus B-tropism , respectively [16] , [48] . Variable region C ( amino acids 375–401 ) contains positively selected amino acids 399 and 401 while residue 399 was also implicated in determining N- versus B-tropism [48] . The nature of the length variation at the C-terminus precludes computational analysis for positive selection; nevertheless functional studies [48] provide compelling evidence that this region can also alter restriction specificity . We have previously noted that CA binding restriction factors Fv1 and TRIM5α share certain design features despite lack of sequence similarity [9]; the present study strengthens this analogy . Both factors possess an N-terminal coiled-coil region allowing dimer formation . They also contain other sequences facilitating the formation of higher order multimers . Both contain a C-terminal domain responsible for virus binding that can be substituted with the cellular CA binding cyclophilin A protein to give a fusion protein capable of restricting HIV-1 and other lentiviruses [23] , [54] . We now provide evidence that the CA binding domain of Fv1 , like TRIM5α [24] , [32] , [55] , [56] , appears to comprise multiple variable regions , showing attributes of positive selection , implying virus driven evolution [3] . Further , Fv1 is capable of recognizing multiple genera of retrovirus . It seems possible that the ability to recognize multiple viruses by low affinity binding with avid binding provided by multimerisation [57] represents a common theme in restriction factor design . Further insights into the interaction between virus and restriction factor requires detailed structural information; unfortunately both Fv1 and TRIM5α are relatively recalcitrant to such studies . The origin of Fv1 remains unclear . It is only present in Mus and appears related to the gag gene of the endogenous retrovirus family ERV-L [18] , [47] . This suggests that Fv1 might be derived from an endogenous retrovirus following the loss of both LTRs and pol coding sequences [58] . Interestingly a significant increase in MERV-L copy number took place at around the time of the separation of Mus subgenera [59] , the time when Fv1 became part of the Mus germline . However sequence alignments indicate that Fv1 and MERV-L share only 43% amino acid identity whereas the different genomic MERV-L elements are much more closely related to one another ( <5% nucleic acid divergence ) . BLAST searches of the NCBI non-redundant genome databases reveal no sequences intermediate between Fv1 and MERV-L . This suggests that Fv1 might be derived from an exogenous virus related to ERV-L that has not made its home as an intact ERV , at least not in any species so far sequenced , and may no longer exist in infectious form . As such Fv1 might be the last remnant of an ancient extinct virus , or paleovirus [2] . Unfortunately this inability to identify the proximal precursor for Fv1 prevents us from determining whether or not the original transgene showed restriction activity and , if so , against which virus . The selection and continuing existence of the Fv1 open reading frame implies that it provides an evolutionary advantage , presumably by providing protection against retroviral infection . The observation of multiple restriction specificities suggests that a variety of unknown viruses have contributed to this process . Taken together with frequent genetic changes to inactivate [60] or block MLV receptors [61] , these data imply that multiple virus epidemics have occurred in the course of mouse evolution [62] . One might postulate that at least four significant virus exposures have occurred during Mus evolution ( Figure 8 ) . One took place after the divergence of Nannomys; a second occurred in M . m . caroli; a third in mice in countries surrounding the Mediterranean Sea and a fourth in the Mus musculus subfamily . In turn this prompts the question of how the current properties of a restriction factor reflect the properties of the viruses involved in selection , a question that is as relevant for TRIM5α as for Fv1 . Specifically one might ask whether the ability to restrict one genus of retrovirus reflects prior exposure to that kind of virus . An affirmative answer might resolve the vexed question of whether foamy viruses have deleterious effects on their hosts [63] , possibly as co-pathogens [64] since both Fv1 ( this paper ) and TRIM5α [29] have evolved to see one or more such virus . Alternatively , changes selected by , say , a gammaretrovirus like MLV , might fortuitously result in recognition of a lentivirus like EIAV or a foamy virus like FFV . In light of the shorter generation time of mice compared to primates Fv1 could provide a more useful system for studying evolution of restriction specificity than does TRIM5α . The observation of multiple alleles of Fv1 might also suggest that selection is an ongoing process offering opportunities for experimental analysis . In particular , the evolution of restriction activity against the lentivirus EIAV , which appears to have happened in two different ways in M . m . spretus and M . m . macedonicus as well as the kind and source ( s ) of the virus ( es ) involved would appear worthy of more detailed investigation .
Genomic DNA samples for Mus musculus laboratory mouse strains C57BL/6J , AKR/J , DBA/2J , 129/SvEv , and LG/J , M . m . spretus ( M . spretus ) , M . m . caroli ( Mus caroli ) , M . m . molissinus ( MOLD/Rk ) and M . m . castaneous were purchased from the Jackson Laboratory . Genomic DNA from M . p . platythrix , M . m . cookii , M . m . spicilegus , M . m . spretus , M . m . castaneous and M . m . bactrianus were gifts from Dr . F . Bonhomme ( Laboratoire Genome et Populations , Universite de Montpellier II , CNRS ) . M . n . minutoides genomic DNA was a gift from Dr . B . Mock ( National Cancer Institute , NIH ) , while M . m . famulus and M . m . cervicolor genomic DNA were gifts from Dr . John Coffin ( Tufts University School of Medicine , Boston ) . M . m . terricolor ( dunni ) genomic DNA was prepared from a Mus dunni tail fibroblast ( MDTF ) line [65] using the DNeasy blood and tissue kit ( Qiagen ) . The Fv1 ORF was PCR amplified from mouse genomic DNA using primers PL80 and GT17 ( see Table S1 for primer sequences ) that permit the amplification of a sequence starting from 3056 bp upstream of the start codon of Fv1 to 2684 bp downstream of the start codon . Sequence analysis of this region from in-bred mice identified 2 SacI sites downstream of the PL80 primer-binding site , while GT17 contained a SalI site . The PCR products were hence cloned initially as SacI/SalI fragments into M13 phage and sequenced . Subsequent subcloning of Fv1 ORFs was carried out following amplification with the primers GatewayFv1F and Gateway Fv1rev . The PCR product was used in a second amplification reaction with primers UniversalF and UniversalRev to attach the attB sites to the ends of the fragment . This was then inserted into pDNR221 , which is an entry vector to the Gateway Cloning system , using BP clonase ( Invitrogen ) . Finally , the entry clone was used in a LR reaction with LR clonase to insert the Fv1 ORF into either pLgatewayIRESEYFP or pLgatewaySN to generate retroviral delivery vectors carrying either the EYFP or G418 resistance marker . Details of these different clones as well as the abbreviations used for their designation are summarized in Table 1 . Fv1 open reading frames from M . m . macedonicus , M . n . minutoides , M . n . gratus , M . n . setulosis , M . n . triton and M . p . saxicolor were synthesized chemically ( GENEART , Life Technologies ) based on their published sequences [40] with added attB sites and introduced into pLgatewayIRESEYFP via pDNR221 . These clones are also summarized in Table 2 . Fv1 chimeras were generated by overlapping PCR . Briefly , a 5′ fragment was amplified from one parental sequence while a 3′ fragment was amplified from the other . The two fragments were then combined in a third amplification reaction using forward and reverse primers that annealed to the 5′ and 3′ ends of Fv1 respectively . Internal primer pairs were designed to target regions of identity between the two parental sequences . The sequences of the primers are shown in Table S1 . To generate the Fv1nC series , the 5′ fragments were amplified from Fv1n using TopoFv1F and either C1Rev , C3Rev , C4Rev or C5Rev , while the 3′ fragments were amplified from Fv1caroli ( CAR1 ) using either C1F , C3F , C4F or C5F and Fv1caroliRev . The 2 fragments were joined in a reaction using TopoFv1F and Fv1caroliRev to yield Fv1nC1 , Fv1nC3 , Fv1nC4 and Fv1nC5 . Similarly , the 5′ fragments for the reciprocal series Fv1Cn were amplified from Fv1caroli ( CAR1 ) , using the same primer pairs as the Fv1nC series , while the 3′ fragments were amplified from Fv1n using either C1F , C3F , C4F or C5F and Fv1nRev . These fragments were joined using primer pair TopoFv1F and Fv1nRev , yielding Fv1Cn1 , Fv1Cn3 , Fv1Cn4 and Fv1Cn5 . The 5′ fragments for the Fv1nS series were amplified from Fv1n using TopoFv1F and either S1Rev , S2Rev or S3Rev , while the 3′ fragments were amplified from Fv1spretus ( SPR1 ) using either S1F , S2F or S3F and Fv1spretusRev . The 2 fragments were joined together in a reaction using TopoFv1F and Fv1spretusRev to yield Fv1nS 1 , Fv1nS2 and Fv1nS3 . Similarly , the 5′ fragments for the reciprocal series Fv1Sn were amplified from Fv1spretus ( SPR1 ) , using the same primer pairs as the Fv1nS series , while the 3′ fragments were amplified from Fv1n using either S1F , S2F or S3F and Fv1nRev . These fragments were joined using primer pair TopoFv1F and Fv1nRev , yielding Fv1Sn1 , Fv1Sn2 and Fv1Sn3 . The chimeric fragments were cloned into pENTR/D-TOPO ( Invitrogen ) and verified by sequencing before transferring into the retroviral vector pLgatewayIRESEYFP . The point mutants were generated by site directed mutagenesis using the primer pairs listed in Table S1 . Mutagenesis was carried out in 50 microlitre reactions containing 2 . 5 units of Pfu ultra , 10 ng of template , 0 . 2 mM dNTP and 125 ng each of the forward and reverse primer . The reaction was performed in a thermal cycler at 95°C for 2 minutes followed by 18 cycles of 95°C for 30 seconds , 55°C for 1 minute and 68°C for 9 minutes 30 seconds . The PCR product was then digested with10 units of DpnI ( Roche ) for 1 hour before transforming XL10Gold cells ( Agilent technologies ) . Colonies were screened by restriction digest and the mutations were verified by sequencing . MDTF and 293T cells were maintained in DMEM containing 10% foetal calf serum and 1% penicillin and streptomycin . Viruses were made by the transient transfection of 293T cells as previously described [19] , [51] . Delivery viruses were produced by co-transfecting pcz-VSVG , pHIT60 and a retroviral vector containing Fv1 and either the EYFP or G418 resistance gene . N- , B- and NB-tropic MLV tester viruses were generated by co-transfection of pczVSVG , pczCFG2fEFPf and either pCIGN , pCIGB or pHIT60 respectively , while the NR-tropic viruses were made using a mutagenized form of pCIGN as previously described [16] . EIAV tester viruses were made using pczVSVG , pONY3 . 1 and pONY8 . 4ZCG or pONY4 . 1Z [66] , while PFV , SFV and FFV were produced with pciSFV-1envwt and either pczDWP001 , pcDWS001 or pcDWF003 respectively [29] . HIV-1 tester viruses were generated by co-transfecting pczVSVG with p8 . 91 and pCSGW . MLV and HIV-1 were frozen in aliquots at −80°C while EIAV and foamy viruses were freshly prepared for each experiment . Restriction activity was routinely assayed using transient two colour FACS analyses as described previously [19] , [51] . Briefly , Fv1 was introduced into MDTF cells together with an EYFP marker in a retroviral delivery vector . Three days post-transduction , the cells were challenged with tester viruses carrying the EGFP markers . The cells were then subjected to FACS analyses three days later and the percentages of tester virus positive cells in EYFP ( i . e . Fv1 ) - positive and - negative cells determined and compared . Ratios of less than 0 . 3 were taken as restriction while those that were greater than 0 . 7 were taken to represent no restriction . Numbers between 0 . 3 and 0 . 7 were taken to represent partial restriction . Alternatively , single cell clones stably expressing restricting Fv1s were derived by transducing MDTF cells in 12 well plates with limiting dilutions of retroviral vectors carrying Fv1 and a G418 resistance marker . The cells from each well were transferred to a 10 cm dish and G418 was added to a concentration of 1 mg/ml . Well-separated colonies were picked from the dishes when they appeared 7 to 10 days after antibiotic selection was started . Typically , 6 to 8 colonies were picked for each Fv1 cell line , expanded and tested for restriction before being used for virus titration . To titrate tester viruses , MDTF cells and their derivatives were seeded in 12 well plates at a density of 5×104 cells per well 24 hours prior to infection . Increasing amounts of viruses carrying the EGFP marker were then added to the wells and the percentage of infected cells was determined by FACS 3 days post infection . MDTF cells and their derivatives stably expressing Fv1 were seeded in 6 well plates at a density of 5×105 cells per well 24 hours prior to infection . The cells were transduced at an m . o . i . of 1 with equal amounts of viral vectors that had been pre-treated with 10 units/ml of DNase ( Promega ) for 1 hour at room temperature . The cells were harvested 7 or 18 hours post-infection for quantification of late RT products and 2 LTR circles respectively . Total genomic DNA was extracted using the DNeasy blood and tissue kit ( Qiagen ) and 250 mg or 500 mg was used for quantitative PCR to detect late RT products and 2 LTR circles respectively . Primers and probes directed against EGFP [67] were used for quantifying late RT products from MLV and FFV while those directed against LacZ were used for EIAV . The retroviral vectors fEGFPf and pHIT111 were used as standards for EGFP and LacZ quantification respectively . Primers and probes for the detection of MLV 2 LTR circles have been described previously [68] . In order to detect EIAV and FFV 2 LTR circles , primers and probes that amplified and bound to a fragment spanning the 2LTRs were designed . For EIAV 2 LTR circle detection , EIAV2LTRCF ( 5′ACTCAGATTCTGCGGTCTGAG3′ ) , EIAV2LTRCRev ( 5′ACCCCTCATAAAAACCCCAC3′ ) and EIAV2LTRCprobe ( 5′FAM-CTCAGTCCCTGTCTCTAGTTTGTCTGTTCG-Tamra3′ ) were used while FFV2LTRCF ( 5′CCAGAACTCACATGAGTGGTG3′ ) , FFV2LTRCRev ( 5′CTCATCGTCACTAGATGGCAG3′ ) and FFV2LTRCprobe ( 5′FAM-GAAGGACTAACCTATCCCAGGTATAGGCCG3-Tamra' ) were used for the quantification of FFV 2LTR circles . The primer pairs were used to amplify fragments spanning the 2 LTRs from genomic DNA of EIAV or FFV infected cells . The fragments were cloned into pCR-BluntII-TOPO ( Invitrogen ) to be used as standards . Quantitative PCR was performed in 25 ml reactions using the ABsolute QPCR Rox mix from Abgene with 300 nM of each primer and 200 nM of probe . A programme of 50°C for 2 minutes , 95°C for 15 minutes followed by 40 cycles of 95°C for 15 s and 60°C for 1 minute was employed in the Applied Biosystems 7500 real time PCR system . Trees were generated using the MegAlign programme from the DNASTAR Lasergene package . The distance values were calculated using the Kimura distance formula that takes into account the number of non-gap mismatches and silent substitutions . | We have followed the evolution of the retroviral restriction gene , Fv1 , by functional analysis . We show that Fv1 can recognize and restrict a wider range of retroviruses than previously thought including examples from the gammaretrovirus , lentivirus and foamy virus genera . Nearly every Fv1 tested showed a different pattern of restriction activity . We also identify several hypervariable regions in the coding sequence containing positively selected amino acids that we show to be directly involved in determining restriction specificity . Our results strengthen the analogy between Fv1 and another capsid-binding , retrovirus restriction factor , TRIM5α . Although they share no sequence identity they appear to share a similar design and appear likely to recognise different targets by a mechanism involving multiple weak interactions between a virus-binding domain containing several variable regions and the surface of the viral capsid . We also describe a pattern of constant genetic change , implying that different species of Mus have evolved in the face of ever-changing retroviral threats by viruses of different kinds . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"animal",
"models",
"of",
"infection",
"viral",
"clearance",
"genetic",
"mutation",
"virulence",
"factors",
"and",
"mechanisms",
"viral",
"transmission",
"and",
"infection",
"virology",
"genetics",
"host-pathogen",
"interaction",
"biology",
"microbiology",
"evolutionary",
"biology",
"viral",
"replication",
"evolutionary",
"genetics",
"gene",
"function"
] | 2014 | Evolution of the Retroviral Restriction Gene Fv1: Inhibition of Non-MLV Retroviruses |
Several studies propose an influence of chromatin on pre-mRNA splicing , but it is still unclear how widespread and how direct this phenomenon is . We find here that when assembled in vivo , the U2 snRNP co-purifies with a subset of chromatin-proteins , including histones and remodeling complexes like SWI/SNF . Yet , an unbiased RNAi screen revealed that the outcome of splicing is influenced by a much larger variety of chromatin factors not all associating with the spliceosome . The availability of this broad range of chromatin factors impacting splicing further unveiled their very context specific effect , resulting in either inclusion or skipping , depending on the exon under scrutiny . Finally , a direct assessment of the impact of chromatin on splicing using an in vitro co-transcriptional splicing assay with pre-mRNAs transcribed from a nucleosomal template , demonstrated that chromatin impacts nascent pre-mRNP in their competence for splicing . Altogether , our data show that numerous chromatin factors associated or not with the spliceosome can affect the outcome of splicing , possibly as a function of the local chromatin environment that by default interferes with the efficiency of splicing .
The transcribed region of almost all human genes contains introns that must be excised from the pre-mRNA for the exons to be spliced together . This process provides an opportunity to modify the exon content of the mature mRNA and , as such , must be regarded as a powerful source of RNA diversity . This alternative splicing is highly regulated and largely depends on a number of non-snRNP splicing factors that bind site-specifically to sequences present on the pre-mRNAs . Over the recent years , a series of observations has suggested that splicing is also influenced by histone modifying enzymes , readers of histone modifications , and chromatin remodelers such as SWI/SNF . Here , we will refer to these proteins as chromatin factors . The reasoning behind a possible impact of these factors on splicing is that splicing is mostly co-transcriptional , and therefore potentially influenced by proteins associated with the transcribed template . At least three modifications of histone H3 present inside the coding region of genes , namely tri-methylation of H3 at lysines 9 ( H3K9 ) , 27 ( H3K27 ) , and 36 ( H3K36 ) , were shown to affect the outcome of splicing in mammalian cells through their specific recognition by dedicated chromatin factors [1–4] . A role for intragenic DNA methylation has also been advocated either as a modification interfering with the recruitment of the boundary protein CTCF that in turns affects splicing , or as a booster of H3K9 tri-methylation [5 , 6] . H3K4 tri-methylation , a modification tightly associated with transcription start sites , and CHD1 , a chromatin remodeler able to bind this modification have also been linked to the regulation of splicing [7] . That study showed the first interaction between a chromatin factor ( CHD1 ) and components of the U2 snRNP , and suggested that this snRNP may function as a bridge between chromatin and splicing machineries . Other observations have since given support to that idea . In particular , immuno-purification of the splicing factor PRP40A from HeLa cell nuclear extracts brings down U2 snRNP subunits together with SWI/SNF subunits and several CHD family members [8] . Furthermore , experiments in Schizosaccharomyces pombe have revealed genetic interactions between U2 and SWI/SNF subunits [9] . Finally , the U2 snRNP subunit SF3B1 was shown to interact directly with chromatin [10] , with Polycomb group proteins [11] , and with the WSTF-SNF2h chromatin remodeling complex [12] . The U2 snRNP is composed of the U2 snRNA and numerous proteins , including 7 Sm proteins , U2-A’ , U2-B” , and the components of the SF3A and SF3B complexes . It associates with the lariat branch site near the 3’ end of the intron via base-pairing between the U2 snRNA and the pre-mRNA . This binding is primed by the association of the U1 snRNP to the 5’ end of the pre-mRNA and the binding of SF1 and U2AF to the branch site and the polypyrimidine track , respectively . The U1 and U2 snRNPs together with the pre-mRNA form the A complex or pre-spliceosome . In most cases , the positioning of this complex defines the exon-intron borders or splice-sites that will be used [13] . The A complex then associates with the U4/U5/U6 tri-snRNP , and finally U1 and U4 are evicted to generate an active B complex . This complex catalyzes a first transesterification reaction that cleaves between the upstream exon and the intron . Finally , the splicing reaction is complete by a second transesterification reaction that rejoins the two exons and releases the intron as a lariat [14] . In the present study , we wished to investigate to what extend the U2 snRNP was a pivot in connecting splicing to chromatin . To address this issue , we developed several complementary approaches . First , we captured spliceosomes assembled before the second transesterification reaction and showed the presence within this complex of chromatin and transcription factors . In a second inverse approach , we systematically depleted human tissue culture cells from known chromatin factors and examined the impact on a splicing reporter . Finally , we combined for the first time chromatin , transcription , and splicing in a same in vitro reaction to estimate the direct impact of chromatin on the splicing reaction . Together , our observation documents a direct and extensive impact of chromatin factors on splicing with however an outcome that remains difficult to predict possibly because of the influence of chromatin .
U2 is the only snRNP present in every spliceosome complexe . Therefore , to capture chromatin factors associated with the spliceosome in the course of an in vivo splicing reaction , we developed a new procedure for proteomic analysis of the U2 snRNP . Earlier protocols for the purification of spliceosomes mostly relied on in vitro assembly of the splicing machinery on a tagged reporter RNA . The subsequent capture of the tagged RNA did not result in co-purification of any chromatin factor , possibly because , by design , the approach captured only spliceosomes assembled independently of transcription . Here , to capture in vivo-assembled U2 complexes , we engineered HeLa S3 cells to express a FLAG-V5-tagged version of U2-B” ( FV5-U2-B” ) , a constitutive component of the U2 snRNP ( Fig 1A ) . Immuno-purifications in the absence of any cross-linking showed that the recombinant FV5-U2-B” was incorporated into both the 12S and 17S forms of U2 snRNP ( S1A Fig and S1B Fig ) . Nuclear extracts prepared from the FV5-U2-B”-expressing cells ( NEB” ) also retained full competence for splicing of a 32P-labelled AdML reporter pre-mRNA in vitro ( S1C Fig compare lanes 1–3 and 6–8 ) , and immuno-precipitation of FV5-U2-B” from the in vitro splicing reactions led to enrichment in both un-spliced and spliced AdML reporter RNA ( S1C Fig , lanes 9–10 ) , consistent with the presence of the U2 snRNP in all intermediate complexes of spliceosome assembly . We then set up to examine the protein composition of complexes associated with FV5-U2-B” during the splicing reaction . To accumulate pre-spliceosome ( complex A ) and spliceosome ( complex B ) , NEB” was supplemented with ATPγS , an ATP analog which blocks the splicing reaction before the second transesterification step [15] . To facilitate the tracing of splicing complexes engaged in splicing reactions , NEB” with ATPγS was further incubated for 40 min . at 30°C with 32P-labelled AdML reporter pre-mRNA . This tracing allowed to confirm accumulation of complexes A and B in our experimental conditions ( Fig 1B ) . The A and B complexes assembled in vivo on non-radioactive pre-mRNAs together with those assembled in vitro on the tracer pre-mRNA were resolved from non-specific ribonucleoparticles ( H complex ) by gel-filtration chromatography and used for immuno-purification with anti-FLAG antibody ( Fig 1C ) . As revealed by mass spectrometry , this procedure resulted in isolation of most previously characterized splicing factors ( 187 out of 284 ) , including all the core components of the spliceosome and many regulators of splicing ( S1 Table ) . In addition , the FV5-U2-B”-associated complexes contained a large number of chromatin factors ( Table 1 ) . Importantly , endogenous/tagged U2-B” and two chromatin factors ( CHD4 and SMARCC1 ) co-sedimented with both the H complex and the spliceosome ( S1D Fig ) . The presence of U2 snRNP subunits , splicing factors , and chromatin factors in both the H complex- and the spliceosome-fractions was confirmed by western blotting ( Figs 1D , lanes 2 & 3 and S1E , lanes 1 & 2 ) . Yet , splicing and chromatin factors were efficiently co-immunoprecipitated with FV5-U2-B” only when using the spliceosome-fractions ( Figs 1D , lanes 4 & 5 and S1E , lanes 3 & 4 ) . This indicated that these factors were physically together with the U2 snRNP only in the context of an assembled spliceosome . The Polycomb group protein PHC1 that was present in both the H complex- and spliceosome-fractions but not detected in the mass spectrometry data , was not co-immunoprecipitated with FV5-U2-B” and is shown as a negative control ( S1E ) . As will be further discussed below , the chromatin factors associated with FV5-U2-B” were enriched in subunits of chromatin remodeling complexes such as SWI/SNF , but included also readers and writers of histone modifications , and rather surprisingly histones . Altogether , these data showed that U2-snRNPs incorporated into splicing complexes assembled onto pre-mRNAs during or immediately after their transcription , are physically associated with chromatin factors . We next questioned whether the chromatin factors associated with U2-snRNPs were representative of the population of chromatin factors affecting pre-mRNA splicing . To reach a global insight on chromatin factors relevant for the regulation of splicing , we carried out an unbiased RNAi screen of virtually every known human chromatin factor , using a CD44-based splicing reporter . This ponasterone-inducible reporter translates transcriptional activity into Firefly-luciferase luminescence , while inclusion of CD44 alternative exons v4-v5 allows for in-frame splicing of the Renilla mRNA and thereby produces Renilla-luciferase activity ( Fig 2A ) . This reporter ( v4-v5—ren ) and a variant missing the v4-v5 genomic sequence ( int—ren–no in-frame splicing of Renilla ) were inserted randomly into the genome of 293 EcR cells , a cell line not expressing the endogenous CD44 gene ( S2A Fig ) . Comparative analysis of two clonal cell lines having integrated either v4-v5—ren or the control int—ren construct confirmed that Renilla enzymatic activity was detected only when v4-v5 exons were present in the reporter mRNA ( S2A and S2B Fig , light-grey bars ) . Furthermore , depletion of Sam68 , a regulator of CD44 splicing [16] , led to a decrease in splicing of v4-v5 , which correlated with a decrease in the Renilla over Firefly luminescence ratio ( referred to as the R/F ratio—S2C Fig ) . Thus , splicing of our reporter was regulated in a manner comparable to that expected for exons v4-v5 in the context of the endogenous CD44 gene . At note , our reporter is not insensitive to eventual changes in translation efficiencies . We then challenged the reporter with a library of 1155 siRNAs targeting 375 different genes predicted or known to encode regulators or components of chromatin ( an average of 3 siRNAs per target—Fig 2B and S2 Table ) . To take experimental variations into account , we compared the R/F ratio obtained with each siRNA to that obtained with control siRNAs ( unrelated and Sam68 , as negative and positive controls , respectively ) . Thus , a siRNA was considered a hit when its transfection resulted in a variation of the R/F ratio equal to at least 50% of that observed with Sam68 siRNA . A gene was selected when at least two siRNAs qualified as hits . Using this criterion , we identified 62 chromatin factors potentially influencing splicing , including 11 factors for which the R/F ratio was affected by all three siRNAs ( triple hits—Fig 2C grey case ) . Among the 62 factors , ING2 and MLL4 were previously identified in another siRNA screen for chromatin factors affecting splicing of a reporter construct ( S3 Table ) [17] . The 11 “triple hits” were retested in the conditions of the screen , and we confirmed by RT-qPCR that they all had an impact on splicing of v4-v5 exons into the reporter mRNA ( S2D and S2E Fig ) . Comparing the outcome of the proteomic analysis of the FV5-U2-B”-associated complexes and that of the siRNA screen revealed an overlap essentially limited to histones and chromatin remodeling factors ( Table 1 , lines with black squares ) . In particular , subunits of the BAF complex and its variant Brm-Sin3a-HDAC were present in both screens ( Fig 2D ) . SNF2H and its partner Pole3 in the human Chrac complex were also identified in both screens ( Fig 2E ) , as were several members of the CHD family . In contrast , histone modifying enzymes and transcriptional regulators identified in the siRNA screen were predominantly absent from the MS data . We next wished to assess the actual impact of the identified chromatin factors on splicing of endogenous genes . These validation experiments were carried out on not previously characterized “triple hits” from the siRNA screen and five additional “double hits” selected for their association with chromatin remodeling complexes ( CHD5 , SMARCAD1 , SIN3a , SMARCA5 , and POLE3 ) . Among these 14 factors , 7 were found associated with the U2-snRNP ( CHD7 , POLE3 , SIN3A , HDAC2 , CHD5 , SMARCA5 , BRD7 ) , while the remaining 7 were not ( TAF5 , H2AFx , TADA2beta , HAT1 , PHC1 , RCF1 , and SMARCAD1 ) . Depletion of 8 of these factors with siRNAs resulted in reduced usage of the endogenous CD44 variant exons v4-v5 in HeLa cells , a cell line naturally expressing CD44 ( Figs 3A , S3A and S3B ) . In contrast , depletion of CBX8 , HDAC3 , and NCAPD2 , three chromatin factors identified neither in the U2 snRNP proteomic study nor in the siRNA screen , did not affect usage of the v4-v5 exons ( S3C Fig ) . We next examined the effect of the depletion of the 14 selected chromatin factors on several exons previously described as being particularly sensitive to U2 snRNP activity [18] , reasoning that factors associated with the U2 snRNP were likely to affect the inclusion of these exons ( Figs 3B , S3D and S3E ) . In this series of experiments , depletion of 13 chromatin factors , present or not in the FV5-U2-B”-associated complex , affected inclusion of multiple alternative exons . Only Brd7 was found not to affect any exon . The 13 chromatin factors active on the U2 snRNP-sensitive exons had no effect on constitutive exons in the RAC1 or the PSMG1 genes , indicating that overall splicing activity was not affected ( S3F Fig ) ; neither did their depletion affect expression of hnRNPU , an hnRNP that influence the maturation and activity of the U2 snRNP [18] . We also observed no correlation between the effect of the 13 chromatin factors on splicing and that on expression of several essential splicing regulators , some of which had putative binding sites in the exons we examined ( SRSF1 , SRSF3 , SRSF4 , SRSF5 , SRSF6 , and hnRNPA1 , S3G Fig ) . These experiments showed that an unexpectedly wide range of chromatin factors can impact on the outcome of splicing , although some were not identified by proteomic as physically interacting with the U2 snRNP in spliceosome complexes . Yet , sensitivity of an exon to levels of U2 appears as a good predictor of its sensitivity to chromatin , suggesting that U2 is responsive to chromatin-born information . Finally , we note that depletion of the various chromatin factors resulted in either increased or decreased inclusion depending on the exon under scrutiny , revealing that the impact of chromatin factors varies from one exon to the other , possibly as a function of local levels of chromatin compaction or histone modifications . To gain information on how chromatin may influence splicing , we next developed an in vitro assay ( see Methods and Fig 4A ) , which allowed a side-by-side comparison of splicing performed by ( i ) the spliceosome alone , ( ii ) splicing coupled to transcription , and ( iii ) splicing coupled to both transcription and chromatin decondensation . We chose a reporter construct based on Drosophila Fushi tarazu ( Ftz ) exon 1 and 2 , previously used in an in vitro system coupling RNAPII transcription to spliceosome assembly [19] . This splice reporter was put under the control of a CMV promoter , or a chimeric promoter with five GAL4 binding sites located upstream of a minimal Adenovirus E4 promoter ( Gal4-E4—S4A Fig ) . For the latter , transcriptional activity was achieved by supplementing all in vitro reactions with the chimeric transcriptional activator Gal4-VP16 . Co-transcriptional splicing of the pre-mRNA synthesized from the CMV-Ftz DNA template was more efficient than splicing of an identical pre-synthesized and capped pre-mRNA , consistent with earlier observations ( [19] and Fig 4B , compare lanes 2 and 5 ) . This difference in splicing efficiency was less discernible when the Gal4-E4-Ftz DNA template was used instead ( compare Fig 4B lane 2 and Fig 4C lane 4 ) , possibly reflecting previously described influence of promoter sequences on splicing [20 , 21] . To then evaluate the impact of nucleosomes on our co-transcriptional splicing assay , the DNA template was chromatinized by combining purified recombinant human chromatin assembly complex ACF ( SMARCA5 and BAZ1A ) and histone chaperone NAP-1 ( NAP1L1 ) with purified HeLa core histones in the presence of ATP [22] . The regularity of nucleosome spacing on the DNA template was confirmed by micrococcal nuclease digestion , which revealed protected DNA fragments corresponding to a ladder of mono- , di- and oligo-nucleosomes , mimicking the nucleosome periodicity observed with native chromatin ( S4B Fig ) . Chromatinization of the CMV-Ftz DNA template strongly reduced transcription , making assertion of the splicing efficiency virtually impossible ( Fig 4B , compare lanes 4 , 5 and 6 , 7 ) . Chromatinization also reduced transcription from the Gal4-E4-Ftz DNA template , although less radically , and transcriptional activity was partially recovered ( approx . 50% of that observed on naked DNA ) by supplementing the in vitro reactions with acetyl coenzyme A ( CoA ) , and sodium butyrate ( NaB ) ( Fig 4C , compare lanes 9 and 18 ) . Acetyl CoA , a co-factor of histone acetylases , and sodium butyrate , an inhibitor of histone deacetylases , favor histone acetylation and thereby participate in licensing the chromatin for transcription . Neither Acetyl CoA , nor sodium butyrate , nor Gal4-VP16 affected splicing of the pre-synthesized pre-mRNA ( S4C Fig , lanes 2–6 ) . We also verified that the ratio between extract and either naked or chromatinized template had no effect on splicing . These experiments indicated that levels of transcription did not affect the efficiency of the splicing reaction ( % of splicing ) , and also that increased concentration of chromatin constituents did not have any inhibitory effect on splicing ( S4D Fig ) . From these validation experiments , we concluded that our conditions properly emulated chromatin-decondensation associated with co-transcriptional splicing . Interestingly , our in vitro assay showed that the transcription of a chromatinized template leads to pre-mRNA splicing that is less efficient than that detected using a naked DNA template ( Fig 4C , compare lanes 7–9 and 16–18 , with 30% vs . 10% splicing efficiency ) . This observation is the first evidence for a direct effect of chromatin on splicing efficiency . To gain insight in the mechanism behind this impact of chromatin on splicing efficiency , we investigated whether the effect was co- or post-transcriptional . To that end , in vitro reactions with the Gal4-E4-Ftz minigene were supplemented with α-amanitin after 45 min of transcription and either stopped ( ice ) or incubated for another 75 min at 30°c ( chronogram Fig 4D ) . The α-amanitin blocks RNAPII processivity without directly affecting splicing ( S4C Fig , compare lanes 4 and 6 ) . As expected from the previous experiments , splicing during the first 45 min ( phase of transcription and splicing ) was less efficient when using the chromatinized template ( undetectable vs . 27%—Fig 4E , compare lanes 3 and 7 ) . Interestingly , decreased efficiency of splicing as a consequence of chromatin was also observed post-transcriptionally after the addition of α-amanitin ( phase of just splicing—Fig 4E , compare lanes 4 and 8 ) . Chromatin-dependent reduction in post-transcriptional splicing efficiency was also observed when using two additional reporters where Ftz exons 1 and 2 were separated by exons harboring ( S ) or not ( T ) 3 copies of an SF2-binding sites ( S4E Fig , compare lanes 2 and 6 , and 8 and 12 ) . These constructs where the S sequences lead to full inclusion of the intervening exon , while the T sequences results in its exclusion , were also an opportunity to observe that chromatin is unable to override a decision enforced by SR proteins ( S4E Fig , species a and c ) . Altogether , these experiments indicate that the pre-mRNPs generated from the naked and chromatinized templates were not equally competent for splicing . This strongly suggests that chromatin influences the quality of pre-mRNPs assembled co-transcriptionally , which in turn affects the efficiency of splicing . Yet , our observations also suggest that chromatin is involved only in fine-tuning of splicing , with little impact on the effect of splicing enhancers .
Co-transcriptional removal of introns occurs in the vicinity of other gene expression machineries , including the RNAPII and the chromatin remodeling factors . While the impact of the RNAPII is now well documented , a role for chromatin in the regulation of splicing is sustained mostly by correlative observations , and the mechanisms involved remain unclear . Here , we have provided a comprehensive study of the coupling between chromatin and splicing , and we have established an in vitro system to examine this coupling directly . Although we have at this point examined only a limited number of reporter constructs , our data indicate that transcribing pre-mRNA from a chromatinized template influences splicing efficiency , and we propose that this effect is in part mediated by physical interactions between chromatin factors and the spliceosome . Our RNAi screen identified a surprisingly broad range of factors , rather than a specific subset of chromatin complexes . The screen caught nearly every chromatin factor previously reported to modulate splicing ( SWI/SNF , Cbx3/HP1γ , ZMYND11/BS69 , CHDs… ) , supporting the relevance of the hits . Some of these factors , including Cbx3/HP1γ , and ZMYND11/BS69 have been examined for their genome wide effect on splicing , further suggesting that our hits affect exons beyond those examined during the phase of validation [4 , 23] . These genome-wide studies and others on MBD3 and CHD4 also indicate that these chromatin factors only have minor effects on the expression of splicing factors , including SRSF1 , SRSF3 , SRSF4 , SRSF5 , SRSF6 , and hnRNPA1 [24 , 25] . A reasonable explanation for the diversity of the hits is the presumed heterogeneity of the local levels of chromatin compaction and/or the range of histone modifications surrounding each copy of our integrated splicing reporter , like it has for example been described for the various copies of endogenous histone genes ( The Encode Project Consortium ) . In that sense , our screen may serendipitously have probed a large spectrum of chromatin environments influencing the regulation of splicing . The local influence of chromatin was also illustrated by our validation experiments on endogenous genes . These experiments showed that depending on the exon under scrutiny , a given chromatin factor had a variable effect , favoring either exon inclusion or exclusion in a rather unpredictable manner . This is in agreement with an earlier study showing that in human breast cancer MCF7 cells , the HDAC inhibitor TSA and the DNA methylase inhibitor 5azadC promote the inclusion exon E107 of the SYNE gene , while they induce exclusion of exon E33 of the fibronectin gene [26] . Likewise , in Drosophila S2 cells , depletion of SWI/SNF subunits promotes the use of proximal splice sites at some genes , while it favors distal sites at others [27] . A possible source of heterogeneity in the chromatin of exons may be their degree of proximity with promoters and enhancers , caused by DNA looping [28] . The deciphering of the probably very complex combination of regulatory signals at play at a given locus will be required to meet the challenge of anticipating the per gene impact of a chromatin factor on splicing . Our proteomic approach confirmed that the splicing machinery is physically bound to a subset of chromatin factors when spliceosome complexes are assembled in vivo . Some of these factors were previously connected to splicing , including MORF4L2 ( close homolog MRG15 ) , Cbx3/HP1γ , SMARCA2/BRM , EHMT1 and EHMT2 , EZH2 , and multiple HDACs [2–4 , 29–31] . In several earlier proteomic studies of the splicing machinery , such interactions were not detected , or were limited to a few factors . This is likely rooted in the procedures used for purification as these approaches involved characterization of the splicing machinery assembled de-novo on pre-synthetized reporter RNAs . With such a setup , components normally dispensed during transcription will not be loaded onto the spliceosome . Our procedure based on U2-snRNP anchoring overcomes this limitation and allows for the isolation of both de-novo- and in-vivo-assembled spliceosome complexes . In that sense , it resembles the previously described capture of the PRPF40A-U2 snRNP that revealed the presence of CHD4/8 and several SWI/SNF subunits in addition to splicing factors [8] . Among the 15 remodeling factors present in that complex , 13 were also detected by our approach . The U2 snRNP is one of the best-characterized snRNPs of the spliceosome , and while several versions have been described , corresponding to different maturation stages [32] , it is likely that only the most abundant particles have been characterized so far , excluding those associated with the transcribed chromatin . Historically , both genetic and biochemical studies have considered the snRNPs as essential rather than regulatory components of the spliceosome . Recent studies , however , demonstrated that several alternative splicing events are regulated by the levels of core components of the splicing machinery [18 , 33] . The exons we examined to validate our hits were identified as particularly sensitive to levels of U2-snRNP . We speculate that this snRNP may function as a mediator between the splicing machinery and the local chromatin environment , and that exons sensitive to U2-snRNP activity are also likely to be subject to chromatin effects . Finally , we note that the list of “chromatin factors” physically linked to the spliceosome in our proteomic approach actually included histones . This suggests that these primary building blocks of chromatin may impact on the outcome of splicing , possibly by affecting nucleosome assembly when present in limited supply . Indeed , nucleosomes may be involved in exon definition as suggested by the elevated nucleosome occupancy/positioning observed in exons compared to introns ( for a review , see [34] ) . Nucleosome assembly may also be relevant for RNAPII elongation rate and for the formation of loops connecting alternative exons to promoter-positioned nucleosomes [28 , 35] . In this context , we believe that our in vitro system combining chromatin , transcription , and splicing will provide a powerful tool to unravel the molecular network linking histones to spliceosome components during the course of transcription .
DNA templates containing promoter and reporter were generated by PCR , purified and 40 ng of DNA were added to a 15-μl in vitro transcription/splicing reactions . Assays were performed by mixing 5-μl of HeLa nuclear extract ( NE ) prepared as described [19] , 5-μl of transcription/splicing mix and 5-μl template , then incubation at 30°C . Transcription/splicing mix was assembled for each reaction with 0 . 20 μl 32P-UTP ( 3000 Ci/mmol ) , 0 . 5 μl 25× ATP/CP mix ( 12 . 5 mM ATP , 0 . 5 mM creatine phosphate ( di-Tris salt ) ) , 0 . 5 μl MgCl2 ( 80 mM ) , 0 . 75 μl Hepes-KOH ( 0 . 4 M ) , 0 . 1 μl dNTP ( 1 mM ) , 0 . 5 μl 25× NTP mix ( 0 . 2 mM UTP , 0 . 6 mM GTP , 3 . 75 mM CTP and ATP ) , 0 . 05 μl sodium butyrate ( 400 mM ) , 0 . 05 μl acetyl coenzyme A ( 1 mM ) , and H20 up to 5-μl . To activate transcription of template containing GAL4 promoter , the NE was supplemented with 20 ng of recombinant Gal4-VP16 , while to inhibit transcription 200ng of α-amanitin was added per reaction . The dNTP/NTP mix is not required , but in our hands , it significantly increased the efficiency of transcription and removed some unspecific bands associated with DNA synthesis . In vitro splicing of pre-mRNA templates was carried out like for transcription/splicing assays . To construct the Ftz splicing reporter , a fragment containing exon 1 ( 256 bp ) , intron 1 ( 147 bp ) , and exon 2 ( 186 bp ) was amplified by PCR from the DoF1 plasmid [19] , and inserted HindIII/XbaI in pcDNA3 . 1 ( + ) downstream of a CMV promoter . The GAL4-E4-Ftz constructs were generated by replacement of the CMV promoter between the MluI and HindIII restriction sites . To generate the derivative constructs with exon T or S , a ClaI restriction site was created in the intron of constructs mentioned above and PCR fragments containing the respective DUP exons [36 , 37] bordered by a small part of the intron were cloned into the ClaI restriction site . The DNA templates for transcription/splicing assay were amplified by PCR using the universal primers CTTAGGGTTAGGCGTTTTGCGCTG and CAACTAGAAGGCACAGTCGAGGCTG . The luciferase splicing reporters v4-v5–ren and int-ren were inserted into the ecdysone-inducible vector pI-TK Hygro ( kindly provided by R . Karni , Hebrew University Medical School–Jerusalem , Israel ) between the restriction sites HindIII and XhoI ( ligated cohesively with SalI ) . Cloning of these reporter constructs required multiple steps; in brief , the first two ATG codons of Renilla cDNA were removed , an exogenous intron with or without the v4-v5 genomic part of the human CD44 gene was inserted , and finally an IRES and the Firefly luciferase cDNA were inserted downstream of a Renilla cDNA . The details of each construct are available upon request . The pBabe-FV5-U2-B” construct was generated by inserting a U2-B” cDNA into the pBabe vector downstream of Flag and V5 tags [38] . DNA template were chromatinized as described in [39] using the Chromatin Assembly Kit ( Active Motif ) . The chromatinized DNA was digested with MNase as described in [39] for 0 , 30 , 75 and 150 sec and analyzed by agarose gel electrophoresis/ethidium-bromide staining . The HeLa S3 cell line expressing FV5-U2-B” was generated by viral infection with pBabe-FV5-U2-B” and a clonal cell line stably expressing the tagged protein at a high level was selected by immunofluorescence using anti-V5 antibody ( Invitrogen ) and expanded to prepare nuclear extract . Endogenous PP2Cγ hnRNPA1 and U2-B” levels were estimated by western blotting with monoclonal antibodies 7–53 , 4B10 and 4G3 , respectively [38] . U2 snRNP was immunopurified from fresh NEB” nuclear extract using ANTI-FLAG M2 Affinity Gel , then eluted with the 3× FLAG Peptide ( Sigma-Aldrich ) , and resolved on a 15–35% glycerol gradient as described [40]; the proteins and RNAs were analyzed on SDS-PAGE and urea gels respectively . Immunoprecipitations from in vitro splicing reactions were performed as above , but the enriched RNAs were isolated directly from the beads without elution . To enrich spliceosome complexes , in-vitro-splicing reactions ( 1 ml ) were set up in the presence of ATPγS [15] and by using the AdML pre-mRNA reporter transcribed from the DoA1 plasmid [19] . The reactions were next size-selected on a Sephacryl-S500 gel-filtration column ( approx . 106 to 5 . 106 Da ) to resolve complexes assembled onto the radiolabelled splicing reporter [41] . The pooled fractions corresponding to each complex were incubated with anti-FLAG M2 Affinity Gel and the immunopurified factors were analyzed by mass spectrometry as described [42] . Western-blot analysis was carried out with the following antibodies: anti-SF3a120 , 66 , 60 ( gift from A . Krämer , Department of Cell Biology , University of Geneva , Switzerland ) ; anti-U2AF65 [42]; anti-U2-B” ( mAb 4G3 ) ; anti-eF4A3 [43]; anti-hnRNPC1/C2 ( mab 4F4 ) ; anti-hnRNPA1 ( mAb A1/55 ) ; anti-CBX3 ( mAb 1G6 ) ; anti-CHD4 ( Sigma , WH0001108M1 ) ; anti-SMARCC1 ( Sigma , B5186 ) ; anti-SMARCA4 ( mab1E1 ) ; anti-SMARCA2 ( mab4147 ) ; anti-PHC1 ( Sigma , HPA006973 ) . The RNA samples were prepared as described [44] and the cDNA libraries were synthesized with M-MLV reverse transcriptase using oligo dT priming or random primers . Radioactive PCRs were performed using 5’end-labeled primers . The primers used in this study are available upon request . The siRNA library was acquired from Qiagen ( each siRNA is listed in S2 Table ) . The Human Sam68 siRNA ( CGGATATGATGGATGATAT;[45] and siGlo ( Dharmacon ) were used as controls . 293EcR and HeLa cells were transfected with Lipofectamine RNAiMAX ( Invitrogen ) . Regarding the siRNA screen conditions , 104 293EcR v4-v5-ren cells were transfected with the siRNA ( 25 nM final concentration ) in 96-well plates . 60h after transfection , expression of the reporter was induced with Ponasterone A ( 1 μM final concentration ) for 16 h and then the cells were lysed in the following buffer ( 100 mM Tris ( pH 8 ) , 0 . 5% v/v Nonidet P-40 , 10 mM DTT ) . Renilla and Firefly luciferase activities were measured with a Dual-Glo Luciferase assay system ( Promega ) using a FLUOstar OPTIMA microplate reader ( BMG Labtech ) . Recombinant Gal4-VP16 was produced in bacteria , and then first enriched using a histidine tag , followed by ion-exchange chromatography using a SP-FF Hitrap , and finally gel-filtration chromatography with a Superose 12 column . | Splicing is an RNA editing step allowing to produce multiple transcripts from a single gene . The gene itself is organized in chromatin , associating DNA and multiple proteins . Some proteins regulating the compaction of the chromatin also affect RNA splicing . Yet , it was unclear whether these chromatin proteins were exceptions or whether chromatin very generally affected the outcome of splicing . Here , we show that a subset of chromatin proteins is physically in interaction with the enzyme responsible for RNA splicing . In addition , several chromatin proteins not found directly associated with the splicing machinery were also able to influence RNA splicing , suggesting that chromatin compaction very globally plays a role in splicing . This finding was confirmed using the first in vitro assay combining transcription and splicing in the context of chromatin; this assay showed that assembling DNA with chromatin proteins influences the efficiency of splicing . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"gene",
"regulation",
"rna",
"extraction",
"dna-binding",
"proteins",
"dna",
"transcription",
"epigenetics",
"extraction",
"techniques",
"chromatin",
"research",
"and",
"analysis",
"methods",
"small",
"interfering",
"rnas",
"genome",
"complexity",
"genomics",
"chromosome",
"biology",
"proteins",
"gene",
"expression",
"rna",
"splicing",
"histones",
"biochemistry",
"rna",
"spliceosomes",
"rna",
"processing",
"cell",
"biology",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"non-coding",
"rna",
"introns"
] | 2016 | A Broad Set of Chromatin Factors Influences Splicing |
Ras mediates signaling pathways controlling cell proliferation and development by cycling between GTP- and GDP-bound active and inactive conformational states . Understanding the complete reaction path of this conformational change and its intermediary structures is critical to understanding Ras signaling . We characterize nucleotide-dependent conformational transition using multiple-barrier-crossing accelerated molecular dynamics ( aMD ) simulations . These transitions , achieved for the first time for wild-type Ras , are impossible to observe with classical molecular dynamics ( cMD ) simulations due to the large energetic barrier between end states . Mapping the reaction path onto a conformer plot describing the distribution of the crystallographic structures enabled identification of highly populated intermediate structures . These structures have unique switch orientations ( residues 25–40 and 57–75 ) intermediate between GTP and GDP states , or distinct loop3 ( 46–49 ) , loop7 ( 105–110 ) , and α5 C-terminus ( 159–166 ) conformations distal from the nucleotide-binding site . In addition , these barrier-crossing trajectories predict novel nucleotide-dependent correlated motions , including correlations of α2 ( residues 66–74 ) with α3-loop7 ( 93–110 ) , loop2 ( 26–37 ) with loop10 ( 145–151 ) , and loop3 ( 46–49 ) with α5 ( 152–167 ) . The interconversion between newly identified Ras conformations revealed by this study advances our mechanistic understanding of Ras function . In addition , the pattern of correlated motions provides new evidence for a dynamic linkage between the nucleotide-binding site and the membrane interacting C-terminus critical for the signaling function of Ras . Furthermore , normal mode analysis indicates that the dominant collective motion that occurs during nucleotide-dependent conformational exchange , and captured in aMD ( but absent in cMD ) simulations , is a low-frequency motion intrinsic to the structure .
Ras proteins are guanine nucleotide-dependent conformational switches that couple cell-surface receptors to signaling pathways that mediate cell proliferation , growth and development [1] . Signal propagation through Ras is mediated by a regulated GTPase cycle that induces distinct conformations with different affinities for downstream effectors . The binding of GTP switches Ras to an “active” effector interacting form . Subsequent GTP hydrolysis returns Ras to the “inactive” GDP-bound form . Two types of regulatory proteins enhance the intrinsically low rates of these processes . GTPase activating proteins ( GAPs ) promote GTP hydrolysis , whilst guanine nucleotide exchange factors ( GEFs ) promote GDP release and regeneration of the active GTP-bound state . Mutations that lead to deregulated Ras activity are found in over 25% of human tumors [2] . These autonomously active variants are insensitive to the action of GAPs resulting in uncontrolled cell growth . The current work aims to better understand the basis of these oncogenic transformations by deciphering how Ras changes its structure as it executes its enzymatic cycle and how the fidelity of this process is affected by mutations . Conformational changes and oncogenic mutations are largely concentrated in the vicinity of the nucleotide binding site , including the so-called switch regions SI ( residues 25–40 ) and SII ( residues 57–75 ) . Of particular note are the conserved SI threonine ( residue 35 ) and SII glycine ( residue 60 ) which converge to form hydrogen bonds with the γ-phosphate of GTP ( effectively ‘closing’ the nucleotide binding pocket ) . In the absence of the γ-phosphate ( or a suitable analogue such as aluminium fluoride ( AlF3 ) ) the switch regions display fewer structural contacts to the nucleotide and reside in a more ‘open’ conformation . This observation has been likened to a loaded spring , where release of the γ-phosphate after GTP hydrolysis allows the switch regions to relax into their ‘open’ GDP-bound conformations [3] . Crystallographic and spectroscopic studies have indicated that the switch regions exhibit substantial mobility both within and between different nucleotide states [3]–[10] . However , a detailed sequence of events and the mechanism by which certain mutations affect key dynamic rearrangements remains elusive . In the present study we employ simulation approaches to perform a detailed characterization of the dynamics of nucleotide-dependent conformational transitions . Previous unbiased molecular dynamics ( MD ) simulations were restricted to characterizing fluctuations within individual nucleotide states [11] , [12] . As a result , external biasing forces ( e . g . , via targeted molecular dynamics ) [13] , [14] were required to characterize the conformational exchange between nucleotide states . However , sampling by these methods is biased along narrow channels whose transition states may be of unrealistically high energy ( 100 and 120 kcal/mol compared to estimates of ∼22 kcal/mol [15] ) . These findings highlight the need for new simulation approaches to probe bias-free transitions . Recently we reported the observation of spontaneous nucleotide-dependent transition during unbiased MD simulation of the oncogenically active G12V variant [16] . This study suggested the existence of a lower thermally accessible energetic barrier between inactive and active states of this variant that renders it prone to adopt an active conformational state . Here we extend this work with a multi-scale simulation approach employing classical and accelerated molecular dynamics ( cMD and aMD ) [17] , [18] along with normal mode analysis ( NMA ) to study both wild-type and mutant transitions . Multiple-barrier-crossing aMD trajectories are used to characterize the reaction paths of the transitions . Simulated conformations are evaluated by comparison to the distribution of Ras's crystallographic conformers . The application of aMD allows the observation of nucleotide-dependent conformational transitions for wild-type Ras that are practically impossible to see with cMD . Furthermore , NMA indicates that the dominant collective motion that occurs during nucleotide-dependent conformational exchange , and captured in aMD simulations , is a low-frequency motion intrinsic to the structure . A number of highly populated intermediate conformations are characterized and their relation to available experimental structures discussed . Finally , the pattern of correlated motions in the current simulations reveals nucleotide-dependent differences of possible functional significance for membrane association .
Clustering of trajectory conformers was used to visualize the dominant conformations sampled by each simulation ( Figure 2 and Table 1 ) . The most populated cluster in the wild-type GDP-to-GTP transition ( black in Figure 2E–H and Figure S4 ) corresponds to the lower of the three basins in Figure 1G and overlaps with the dominant conformation sampled during cMD simulations of the same system ( Figure 1E ) . Its overall structure is intermediate between GDP and GTP states ( RMSD of 1 . 3 Å from both GDP and GTP representatives , PDB codes 4q21 and 1qra respectively ) . Members of this cluster have an intermediate α2 orientation and a closed active site loop2-SI and loop4-SII that more closely resembles the GTP configuration . The second most populated cluster ( yellow in Figure 2E–H ) has more distinctive GTP like characteristics ( RMSD of 1 Å from GTP and 1 . 6 Å from GDP representatives ) including a closed loop2-SI and loop4-SII active site and a reoriented GTP-like α2 helix . Interestingly , the PC projection ( Figure 2G ) and RMSD values ( minimum value 0 . 4 Å ) indicate that members of cluster 2 closely resemble the crystallographic GTP-bound A59G structure ( PDB code 1lf0 ) . This structure has been suggested previously to be an intermediate and is characterized by a GTP-like SII conformation and a SI conformation that has undergone partial transition with the side-chain of Y32 adopting an orientation that is intermediate between that in wild-type GDP and GTP crystallographic structures [6] , [7] , [16] ( see Figure S2 ) . The third cluster ( green in Figure 2E–H ) is again equidistant from GDP and GTP states ( RMSD of 1 . 3 to 1 . 4 Å ) and , similar to cluster 1 , is characterized by an intermediate α2 orientation . However , unlike cluster 1 , loop2-SI and loop4-SII resemble the open GDP conformation . Clusters 4 and 5 are closer to the GDP conformation ( with RMSDs of 1 and 0 . 3 Å ) than to the GTP conformation ( with RMSDs of 1 . 4 and 1 . 7 Å ) . However , cluster 5 has distinct conformations of loop3 and the C-terminal portion of α5 . The GTP-to-GDP ( backward ) transition sampled a number of distinct conformational states ( Figure 1D and Figure 2A–D ) . Clusters 5 and 2 have a similar loop2-SI conformation and resemble the starting GTP state ( RMSD values of 0 . 9 and 0 . 8 Å , respectively ) . A major difference between clusters 2 and 5 is the orientation of loop3 . Furthermore , conformations in cluster 2 project to a similar region of PC space to those in the second cluster of the forward GDP-to-GTP transition described above . The remaining clusters ( black , green and pink in Figure 2A–D ) have a strikingly open loop2-SI conformation , similar to that observed in crystal structures 1×cm , 1bkd , 1nvv , chain R of 1nvu , chain R of 1nvx , and chain B of 1×d2 . Cluster 3 most closely resembles the GDP state both in terms of its PC projection ( Figure 2C ) and RMS distance ( 1 . 1 Å ) . Clusters 1 and 4 project close to the GTP-bound crystallographic structures in which Y32 has been replaced by a Cys-chromophore . However , RMSD measurements of over 1 . 4 Å indicate features distinct from Y32 mutants and wild-type GDP/GTP states . Taken together , these results suggest that in addition to several distinct conformations that were not observed previously , both the forward and backward transitions pass through a common intermediate similar to that captured in the A59G crystal structure . The temporal evolution of cluster membership in each trajectory ( Figure 2D and 2H ) indicates that the first 4 to 5 ns remain relatively close to the starting structure ( with conformations classified as cluster 5 or 4 ) . This is followed by multiple transitions between states with periods of sampling where one conformation predominates . For example , conformations from the 10 to 25 ns period of the GDP-to-GTP trajectory reside predominantly in the GTP-like cluster 2 . Similarly , conformations from the 33 to 42 ns period of the GTP-to-GDP trajectory are classified as the more GDP-like cluster 3 . These single cluster blocks are interspersed with periods of rapid interconversion between clusters , for example the 30 to 35 ns portion of the GDP-to-GTP trajectory ( Figure 2H ) . Interestingly , the GDP-to-GTP trajectory evolves toward a GTP-like cluster 2 conformation and then transitions back to a intermediate cluster 3 conformation before returning back via clusters 4 and 5 ( see Videos S1 and S2 ) . This behavior is consistent with the presence of a higher barrier between cluster 2 and the main crystallographic GTP cluster , which we fail to cross , than exists between the intermediate conformations contained within our clusters . The high level of inteconversion between these clusters also suggests that they are energetically relatively close to one another . Analysis of the calculated structures indicates that certain side chain reorientations , diagnostic of GTP and GDP crystallographic states [6] , [7] , [13] , [16] , have only been partially realized in our simulations . We speculate that while the topology of the backbone provides the conformational blue print , specific side-chain interactions are required to stabilize the canonical GTP and GDP states . The apparent barrier between cluster 2 and the main crystallographic GTP cluster would therefore result from the energetic cost of reorganizing these side chains . A typical example is Y32 , whose orientation in the simulated intermediates is neither fully solvent exposed nor hydrogen bonded with Y40 as in the main GTP and GDP crystallographic clusters respectively [16] . Rather we note its similarity to that found in the crystal structures 1gnr and 1gnq . To examine whether the motions of one residue are related to the motions of another ( distant ) residue , the correlation of the displacements of all residue pairs were determined ( Figure 3 ) . As expected , the strongest positive correlations exist between covalently bonded residues and those residing within secondary structure elements ( see Figures S5 and S6 for reference contact maps ) . Moving up the diagonal , the first area of notable correlations corresponds to helix α1 ( residues 16 to 25 ) with loop2 ( residues 26 to 40 ) . The next area of significant correlation corresponds to the β2-loop3-β3 region ( residues 38 to 57 ) . The consistent appearance of correlated motions for these regions in each simulation and the cross-correlation ( off-diagonal peak ) with β1 ( residues 2 to 10 ) highlights the subdomain-like structure and dynamics of these three N-terminal strands . The remaining strands , β4 to β6 , display consistent positive cross-correlations with each other but not with strands β1 to β3 . Note the off-diagonal peaks for β4 ( residues 77 to 83 ) with β5 ( residues 111 to 115 ) and β5 with β6 ( residues 141 to 144 ) . Moving further up the diagonal the next major correlations correspond to the SII region , encompassing loop4 and helix α2 ( residues 58 to 74 ) . Perhaps the most interesting feature of the plot is the pattern of correlation between α2 and α3-loop7 ( residues 66 to 74 and 93 to 110 ) . This feature is most evident in GTP-bound simulations and is largely absent in GDP-bound simulations . This pattern is particularly noteworthy as GDP-to-GTP aMD simulations exhibit these correlations only in portions of the trajectory that reside in a GTP like conformation ( i . e . after the transition from GDP to GTP , see Figure 2G and 2H ) . It appears that the large rearrangement of helix α2 during the transition brings it into closer register with α3 , hence facilitating the correlated motions of these regions . Furthermore , the correlations of these regions in the GTP-to-GDP aMD simulation reduce gradually as the conformation of α2 evolves toward a more GDP like state . These data thus show a novel GTP-dependent correlated motion in Ras that has functional implications ( see below ) . Additional off-diagonal peaks include loop2 with loop10 ( residues 26 to 37 and 145 to 151 ) and loop3 with α5 ( residues 46 to 49 and 152 to 167 ) . As discussed below loop3 and α5 are connected via several salt bridges whilst both loop2 and 10 directly interact with the bound nucleotide . These newly identified coupled motions suggest a dynamic linkage between the N-terminal nucleotide-binding subdomain and the C-terminal subdomain whose downstream residues are responsible for membrane binding . The dissection of the catalytic domain into two lobes or subdomains based on the correlated motions of the central β-strands is consistent with the localized nature of sequence variation between Ras isoforms . As previously noted [16] , lobe 1 ( residues 1–86 ) is strictly conserved in sequence and encompasses the P-loop and the switch regions; whilst lobe 2 ( residues 87–171 ) contains amino acid variations that define functionally distinct H , N and K-ras isoforms [16] . As isoform-specific properties include differences in nucleotide-state dependent membrane localization [19]–[21] , the segregation of both sequence variation and correlated motions implies that communication between lobes is likely to be of functional significance . The covalent connection between lobes is made by helix α2 of the SII region , which is the major dynamic element of the Ras structure . The current cross-correlation analysis indicates the existence of three additional non-covalent communication routes between lobes including loop3 to α5 , loop2 to loop10 and α2 to α3-loop7 . We speculate that residues at each of these sites may be important for nucleotide-dependent modulation of membrane attachment and lateral segregation by linking the switching apparatus to the membrane interaction apparatus . Indeed , alanine substitution of loop3 residues D47 and E49 produced a variant that is hyperactive in MAPK-signaling [22] . These loop3 residues , together with their α5 salt bridge partners ( R161/R164 ) , have been shown to modulate the nucleotide-dependent membrane association of H-ras [22] , [23] . The other regions highlighted in the current study have thus far received little attention but likely warrant further investigation . In an effort to further understand the physical basis of the observed motions upon nucleotide exchange , we analyzed available structures with a simplified elastic-network normal mode method [24] . The elastic network approach has the advantage that a single model , expressed in terms of Cα coordinates , leads to an objective expression of possible protein dynamics in terms of a superposition of collective normal mode coordinates [25] . Consistent with previous studies on a range of systems [26] , we note that only the “open” wild type GDP conformation yielded large overlap values between the crystal structure PCs and the low-frequency normal modes ( overlap between NMA mode 1 and X-ray PC 1 of 0 . 57 ) . Furthermore , the structural mobility predicted by NMA is very similar to that obtained from aMD simulations ( overlap between GDP-to-GTP trajectory PC 1 and X-ray PC 1 of 0 . 69; Figure 4 and Table S2 ) . This result implies that low-frequency global motions that are intrinsic to the open structure likely facilitate the observed conformational transitions to the closed GTP state . In other words , the fact that low frequency normal modes qualitatively capture the differences between available crystal structure conformations and have high overlap with the eigenvectors obtained from aMD simulations suggests that nucleotide-dependent dynamics is facilitated by the low frequency , global motions that are intrinsic to the structure . The aMD results discussed above indicate that the nature of the bound nucleotide attenuates these intrinsic motions . We have characterized the spontaneous transition between nucleotide-dependent conformational states of wild-type Ras with cMD , aMD and NMA . These functionally important transitions , achieved for the first time for wild-type Ras , are practically impossible to observe with cMD . Furthermore , NMA indicates that the dominant collective motion that occurs during these transitions is a low-frequency motion intrinsic to the structure . Mapping the reaction path sampled by aMD onto a PCA basis set derived from the distribution of crystallographic structures enabled identification of intermediate structures with unique switch orientations and/or distinct loop3 , loop7 and α5 C-terminus conformations . Intriguingly , several of the highly populated intermediates have a close correspondence to known G59A and Y32C crystallographic conformers , both of which have been suggested to be intermediate structures [6] , [16] . The emergence of these conformations along with additional novel intermediates highlights the utility of aMD simulations to reliability sample conformational transitions . Furthermore , the current results imply that the G59A and Y32C variant conformations are accessible to wild-type Ras and that these mutations result in perturbations that localize the average structure at these intermediate positions . The functional relevance of these intermediates is reinforced by kinetic studies of G59A that indicated a reduced rate of nucleotide exchange; this has been linked to the need for tight nucleotide coordination during structural changes [6] , [16] . It will be interesting to see if our newly identified intermediates , some of which differ between forward and backward transitions , also have distinct kinetic behaviors related to nucleotide exchange and phosphate release . The pattern of correlated motions revealed by these simulations predicts novel nucleotide-dependent motions of potential significance in the signaling function of Ras . These include correlations of α2 with α3-loop7 , loop2 with loop10 and loop3 with α5 . Such dynamic linkages between the switching apparatus and the membrane interacting C-terminal region leads us to speculate that residues at each of these sites may be important for nucleotide-dependent modulation of membrane attachment . This is supported by recent experimental evidence for the role of loop3 residues D47 and E49 and α5 residues R161 and R164 in modulating the nucleotide-dependent membrane association of Ras [22] , [23] . Finally , low frequency normal modes qualitatively capture the differences between available crystal structure conformations and have high overlap with the eigenvectors obtained from aMD simulations . This result combined with aMD observations suggests that nucleotide-dependent dynamics is facilitated by low frequency , global motions that are intrinsic to the structure and that the nature of the bound nucleotide serves to attenuate these intrinsic low-frequency motions . Furthermore , the significant similarities of aMD , NMA and crystal structure PCA motions highlight the robustness of the observed motions . We believe that the current advanced simulation and analysis approach is equally applicable to a large number of structurally similar but functionally diverse P-loop NTPases such as kinesin and myosin . Such studies should uncover detailed dynamic behavior and help inform us about general principles and mechanisms underlying nucleotide-dependent conformational changes .
All MD simulations were performed using periodic boundary conditions , TIP3P water and charge-neutralizing counter ions , with full particle-mesh Ewald electrostatics . Operational parameters included a 2fs time step and a 10Å cutoff for the truncation of VDW non-bonded interactions . Constant volume heating ( to 300 K ) was performed over 10ps , followed by constant temperature ( 300 K ) , constant pressure ( 1atm ) equilibration for an additional 200ps . Finally , constant pressure constant temperature production dynamics was performed with both classical and accelerated MD implementations . The SHAKE algorithm was used to constrain all covalent bonds involving hydrogen atoms . Accelerated MD ( aMD ) extends the accessible time scale of conventional MD simulations by altering the underlying potential energy surface of the system under study . Acceleration stems from the addition of a non-negative boost potential that raises the energy within basins [18] . Hence a trajectory propagated on this modified surface makes transitions from state to state with an accelerated rate . Furthermore , canonical ensemble averages of the system can be obtained by reweighing each point on the modified potential by the strength of the Boltzmann factor of the bias energy at that particular point [18] . In the current study we apply the dual boost approach and corresponding potential developed previously [17] , [18] . Starting structures and standard operational parameters were identical to those used for cMD . The energy level , E , below which the boost is applied and tuning parameter , α , that modulates the depth and local roughness of basins in the modified potential were based on an earlier work [17] . We employed the coarse-grained AD-ENM normal mode analysis approach developed by Zheng et al . [24] . AD-ENM implements a single-parameter Hookean potential , which has previously been shown to yield low-frequency normal modes that are in good agreement with those obtained from more detailed , empirical , force fields . For further details see [24] , [25] PCA was employed to aid the interpretation of interconformer relationships . We utilized the previously reported PC basis set obtained from analysis of available Ras crystal structures [16] . This basis set gives a clear separation of nucleotide-dependent conformational states . Projecting the Ras crystal structures and snapshots from MD trajectories into the sub-space defined by the largest PCs ( along which the crystal structure variance is largest ) results in a lower dimensional representation of the structural dataset ( see Figure 2 for details ) . The resulting low-dimensional ‘conformer plots’ , succinctly reveal the nature of conformational sampling during simulations [30] . PCA was carried out on the individual trajectories using the same Cα atoms that were used in the analysis of the crystal structures . Conformer superposition was also based on the “core positions” obtained from crystal structure analysis [16] . Distances between instantaneous trajectory conformations and the centroids of the main GTP and GDP clusters , reported as inserts in Figure 1 , were calculated as the Euclidean distance between projected points in five dimensional PC space . Note that five PCs account for over 81% of the variance in the original distribution and produce a more succinct distance measure than the examination of average all-atom distances . This metric aids interpretation of an otherwise noisy signal as it is derived primarily from the concerted displacement of the switch regions comprising the secondary structure elements loop2 and loop4-α2 ( residues 31 to 37 and 59 to 72 ) . Structures from aMD simulations underwent average-linkage hierarchical clustering according to their pairwise RMSD distance matrix . Inspection of the resulting dendogram was used to partition structures into five dominant groups ( ranked according to their populations ) . The closest structure to the average structure from each cluster , in terms of RMSD , was chosen as a representative for projection onto the PCA basis set described above . To identify protein segments with correlated atomic motions the cross-correlation coefficient , Cij , for the displacement of all Cα atom pairs , i and j , was calculatedwhere Δri is the displacement from the mean position of the ith atom determined from all configurations in the trajectory segment being analyzed ( see [32] and [33] for further details ) . | The Ras family of enzymes mediate signaling pathways controlling cell proliferation and development by cycling between active and inactive conformational states . Mutations that affect the ability to switch between states are associated with a variety of cancers . However , details of how the structural changes occur and how mutations affect the fidelity of this process remain to be determined . Here we employ an advanced computational technique , termed accelerated molecular dynamics , to characterize structural transitions and identify novel highly populated transient conformations . Several spatially distant structural regions were found to undergo correlated motions , highlighting a dynamic linkage between the sites of enzymatic reaction and the membrane-interacting C-terminus . In addition , our results indicate that the major motion occurring during the conformational exchange is a low-frequency motion intrinsic to the structure . Hence , features of the characterized transitions likely apply to a large number of structurally similar but functionally diverse nucleotide triphosphatases . These results provide fresh insights into how oncogenic mutations might modulate conformational transitions in Ras . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Methods"
] | [
"biophysics/theory",
"and",
"simulation",
"biochemistry/cell",
"signaling",
"and",
"trafficking",
"structures",
"biochemistry/bioinformatics",
"biochemistry/theory",
"and",
"simulation",
"computational",
"biology/molecular",
"dynamics",
"biophysics/cell",
"signaling",
"and",
"trafficking",
"structures"
] | 2009 | Ras Conformational Switching: Simulating Nucleotide-Dependent Conformational Transitions with Accelerated Molecular Dynamics |
Condensin-mediated chromosome condensation is essential for genome stability upon cell division . Genetic studies have indicated that the association of condensin with chromatin is intimately linked to gene transcription , but what transcription-associated feature ( s ) direct ( s ) the accumulation of condensin remains unclear . Here we show in fission yeast that condensin becomes strikingly enriched at RNA Pol III-transcribed genes when Swd2 . 2 and Sen1 , two factors involved in the transcription process , are simultaneously deleted . Sen1 is an ATP-dependent helicase whose orthologue in Saccharomyces cerevisiae contributes both to terminate transcription of some RNA Pol II transcripts and to antagonize the formation of DNA:RNA hybrids in the genome . Using two independent mapping techniques , we show that DNA:RNA hybrids form in abundance at Pol III-transcribed genes in fission yeast but we demonstrate that they are unlikely to faciliate the recruitment of condensin . Instead , we show that Sen1 forms a stable and abundant complex with RNA Pol III and that Swd2 . 2 and Sen1 antagonize both the interaction of RNA Pol III with chromatin and RNA Pol III-dependent transcription . When Swd2 . 2 and Sen1 are lacking , the increased concentration of RNA Pol III and condensin at Pol III-transcribed genes is accompanied by the accumulation of topoisomerase I and II and by local nucleosome depletion , suggesting that Pol III-transcribed genes suffer topological stress . We provide evidence that this topological stress contributes to recruit and/or stabilize condensin at Pol III-transcribed genes in the absence of Swd2 . 2 and Sen1 . Our data challenge the idea that a processive RNA polymerase hinders the binding of condensin and suggest that transcription-associated topological stress could in some circumstances facilitate the association of condensin .
Mitotic chromosome condensation is essential for genome integrity . When defective , chromosomes often remain entangled and fail to segregate properly in anaphase . A key driver of chromosome condensation is the highly conserved condensin complex . Condensin is made of five sub-units ( SMC2Cut14 , SMC4Cut3 , CAP-D2Cnd1 , CAP-GCnd3 and CAP-HCnd2 , name of the human protein followed by its name in fission yeast ) and it is one of the main components of mitotic chromosomes [1] . In vitro , purified condensin can introduce positive supercoils into a relaxed plasmid in the presence of topoisomerase I [2] , [3] . These observations support the idea that condensin shapes mitotic chromosomes by changing the topology of chromatin around its binding sites . However , the mechanisms underlying the association of condensin with chromatin remain poorly understood ( reviewed in [4] ) . Several studies have illustrated the paradoxical relationships linking gene transcription and the localization of condensin . From pro- to eukaryotes , condensin is preferentially enriched at highly transcribed genes [5] , [6] , [7] , [8] , suggesting that some highly conserved transcription-associated feature ( s ) that predate ( s ) the appearance of nucleosomes help to recruit condensin . However , experiments in yeast indicated that RNA polymerases must be silenced before condensin can bind , at least at repetitive sequences such as the rDNA or the sub-telomeres [9] , [10] . These somewhat contradictory observations could potentially be reconciled if one hypothesizes that a by-product of the transcription process facilitates the recruitment of condensin . In this study , we have considered that such a by-product could be R-Loops or transcription-associated topological stress . R-Loops result from the formation of stable DNA:RNA hybrids in the genome . As a consequence of the hybridization of the RNA to the template , the non-transcribed strand of the DNA remains single-stranded ( reviewed in [11] ) . Interestingly , the hinge domain of the Smc2/Smc4 heterodimer in condensin shows high affinity in vitro for single-stranded DNA [12] , [13] . Moreover , a recent study proposed that chromatin is less accessible to restriction enzymes in mutants where R-Loops accumulate , consistent with the idea that R-Loop formation favours chromatin compaction [14] . Interestingly , fission yeast condensin can disassemble DNA:RNA hybrids in vitro [15] and its chicken counterpart localizes to CpG islands [6] , which constitute major R-Loop forming regions in the genome [16] . Taken together , these observations support the idea that R-Loops and condensin could interact functionally in vivo [14] . According to the twin supercoiled domain model , high rates of transcription induce positive supercoiling of the chromatin in front of the elongating polymerase , whilst negative supercoiling accumulate upstream of the polymerase [17] . As such , highly expressed genes represent regions of the genomes that accumulate topological stress . As confirmed in vivo recently , this stress is monitored by topoisomerase I and topoisomerase II [18] , [19] , [20] . Interestingly , in vitro assays have indicated that condensin binds preferentially to positively supercoiled plasmids in the presence of ATP [21] . Whether or not this transcription-associated topological stress contributes to the binding of condensin in vivo has not been addressed . In order to clarify the functional relationships between transcription and chromosome condensation , we recently carried out a genetic screen in fission yeast to identify deletions of transcription-associated factors that would rescue a condensin deficiency [22] . For this , we isolated loss-of-function mutations that could rescue the thermo-sensitivity of the condensin mutant cut3-477 [23] . Two of the mutations we isolated were the deletions of swd2 . 2 ( swd2 . 2Δ ) and sen1 ( sen1Δ ) [22] . Swd2 . 2 is a non-essential component of the Cleavage and Polyadenylation Factor ( CPF ) , the complex responsible for 3′end maturation of RNA Pol II transcripts in yeast ( reviewed in [24] ) , where it acts to maintain the proper levels of CPF-associated phosphatases [22] . Fission yeast Sen1 is the homologue of human Senataxin and has been shown to unwind DNA:RNA hybrids in vitro [25] . Budding yeast Sen1 is involved in transcription termination [26] but its role in fission yeast has not been characterized . Here we show that both factors act directly at Pol III-transcribed genes to limit the association of condensin and the accumulation of topological stress . Furthermore , topological stress at Pol III-transcribed genes facilitates the association of condensin when Swd2 . 2 and Sen1 are missing .
On their own , the deletions of swd2 . 2 ( swd2 . 2Δ ) and sen1 ( sen1Δ ) partly restored growth of cut3-477 cells at the restrictive temperature ( Figure 1A ) and reduced the proportion of anaphase cells displaying chromosome segregation defects ( Figure 1B ) . Combining both deletions ( sen1Δswd2 . 2Δ ) resulted in a stronger suppressor effect ( Figure 1AB ) . The double mutant sen1Δswd2 . 2Δ also suppressed the other condensin mutant cut14-208 ( Figure S1 ) . Strikingly , Chromatin Immunoprecipitation ( ChIP ) analysis in cycling cell populations showed that the localization of condensin was altered at specific loci when Swd2 . 2 and Sen1 were both missing: its recruitment increased significantly at genes transcribed by RNA Pol III ( Gln . 04 , Met . 07 , Ser . 13 , Pro . 09 , Tyr . 04 , Gly . 05 , 5S rRNA , Arg . 04 on Figure 1C ) , whereas it was significantly reduced at the rDNA arrays ( 18S&Rfb2 ) . The binding of condensin remained unaffected at kinetochores ( cnt1 ) or at highly transcribed Pol II genes ( Act1 , Adh1 , Fba1 and SPAC27E2 . 11c ) . The sequences of all the primers used in this study are available on Table S1 . The mitotic indexes of both cell populations ( swd2 . 2+sen1+ and swd2 . 2Δsen1Δ ) were comparable ( Figure 1D ) , ruling out that the changes in the association of condensin are due to indirect , cell-cycle defects . These data established that Sen1 and Swd2 . 2 act to limit the localization of condensin at Pol III-transcribed genes . The reasons why the association of condensin at the rDNA arrays is reduced in the absence of Swd2 . 2 and Sen1 will be explained elsewhere . We found previously that Swd2 . 2 associates with Pol III-transcribed genes and that lack of Swd2 . 2 restored the localization of condensin at Pol III-transcribed genes in the condensin-deficient mutant cut3-477 [22] . Here , we show that Sen1 is also significantly enriched at Pol III-transcribed genes and that its binding is independent of Swd2 . 2 ( Figure 2A ) . Furthermore , affinity purification of Sen1 followed by mass-spectrometry analysis of its associated proteins identified most sub-units of the RNA Pol III complex as its most stable binding partners ( Table S2 ) . We confirmed this interaction by showing that the RNA Pol III sub-unit Rpc25 co-precipitates with Sen1 ( Figure 2B ) . Note however that Sen1 did not co-precipitate with Sfc6 , a sub-unit of TFIIIC ( Figure S2 ) , a complex required for the association of RNA Pol III with chromatin [27] . ChIP analysis showed that the association of Rpc25 with chromatin was significantly increased in the absence of Sen1 ( Figure S3 ) or in swd2 . 2Δsen1Δ cells ( Figure 2C&D ) . In swd2 . 2Δsen1Δ cells , the stabilization of RNA Pol III on chromatin was associated with an increase in the steady-state level of tRNAs , as detected by RT-qPCR analysis ( Figure 2E ) . Taken together , these experiments concur to show that Swd2 . 2 and Sen1 play a direct role at Pol III-transcribed genes , where they limit the association of RNA Pol III and the accumulation of transcripts . These results show that the accumulation of condensin at Pol III-transcribed genes in swd2 . 2Δsen1Δ cells is concomitant with an enhanced transcriptional activity . It was recently argued that budding yeast Sen1 limits the accumulation of DNA:RNA hybrids , including at Pol III-transcribed genes [28] . Fission yeast Sen1 similarly was shown to display a DNA:RNA helicase activity in vitro [25] . These observations and the additional arguments detailed in the introduction prompted us to test the possibility that R-Loops could represent a transcription by-product facilitating the association of condensin with chromatin . We speculated that lack of Sen1 and Swd2 . 2 could result in the accumulation of R-Loops at Pol III-transcribed genes where they might contribute to increase the association of condensin . To establish whether or not R-Loops form at Pol III-transcribed genes in fission yeast , we first monitored by ChIP the chromatin association of RNase H1 , one of the endogenous enzymes known to disassemble R-Loops . More specifically , we introduced at the endogenous locus a point mutation ( D129N ) in the fission yeast RNase H1 ( Rnh1 ) , because the same mutation was shown to weaken the catalytic activity of human RNase H1 [29] . Consistent with this , the D129N mutation did stabilize the interaction of Rnh1 with Pol III-transcribed genes ( Figure 3A ) . Furthermore , the interaction of Rnh1D129N with Pol III-transcribed genes was lost upon over-expression in vivo of RnhA , the RNase H1 enzyme from E . coli ( Figure 3B ) . Upon over-expression , RnhA itself did not stably associate with Pol III-transcribed genes ( Figure S4 ) , showing that the loss of Rnh1D129N from Pol III-transcribed genes upon over-expression of RnhA cannot be explained by its mere replacement by bacterial RnhA . Finally , Figure S5 shows that the association of Rnh1D129N with the rDNA repeats increased significantly in the absence of topoisomerase I ( top1Δ ) , consistent with the observations reported previously that lack of Top1 triggers the accumulation of R-Loops at rDNA in budding yeast [30] . This confirmed that Rnh1D129N was able to detect significant changes in R-Loop accumulation . Taken together , these data show that ChIP with Rnh1D129N is a reliable way to identify R-Loop forming regions in fission yeast . We sought to confirm the formation of R-Loops at genes transcribed by RNA Pol III using another approach . A method that is commonly used to map R-Loop forming regions in yeast is to perform ChIP using the S9 . 6 antibody because of its high affinity for DNA:RNA hybrids [31] . ChIP requires formaldehyde cross-linking followed by sonication of the chromatin . We found that the ability of S9 . 6 to detect R-Loops generated after transcription in vitro was greatly diminished both by formaldehyde cross-linking and by sonication ( Figure S6 ) . We do not know at this stage whether this is because R-Loops are partly destroyed by these treatments or because these treatments reduce the affinity of the antibody for R-Loops . To circumvent these issues , we extracted genomic DNA from unfixed cells , digested soluble RNA using RNase A and sheared the DNA using a cocktail of restriction enzymes ( see Methods ) . Dot blot analysis using the S9 . 6 antibody confirmed that our procedure largely preserved R-Loops ( Figure 3C ) . We then performed DNA:RNA immuno-precipitation ( DRIP ) using the S9 . 6 antibody in stringent conditions , in the presence of 500 mM NaCl . As expected , the DRIP signal at 18S , the canonical R-Loop forming region within the rDNA repeats [30] , increased significantly in the absence of RNase H1 and RNase H2 ( rnh1Δrnh201Δ cells ) and disappeared almost entirely upon treatment of the genomic DNA with commercial RNase H ( Figure 3D ) . On the contrary , the DRIP signal detected at a non-transcribed region NT ( chr I , 3009300-3009500 , [32] ) remained low both in rnh1Δrnh201Δ cells and upon treatment with RNase H . Those controls demonstrated that the signals we detected using DRIP were specific . In agreement with the results obtained using ChIP of Rnh1D129N as a reporter for the presence of R-Loops , we detected strong DRIP signals at Pol III-transcribed genes in the absence of RNase H1 and RNase H2 ( Figure 3D ) . In conclusion , the two methods we have set up to map R-Loop forming regions establish that R-Loops are a prominent feature of Pol III-transcribed genes in fission yeast . Using ChIP of Rnh1D129N , we established that R-Loops accumulate to similar levels at Pol III-transcribed genes in cycling cells ( >90% of interphase cells ) and in cells synchronized in early mitotis ( Figure S7A ) . Consistent with this , ChIP established that the association of RNA Pol III with chromatin is largely maintained in mitosis ( Figure S7B ) . Taken together these experiments support the idea that transcription at Pol III-transcribed genes is maintained in mitosis , at a time when condensin is loaded on chromosomes in fission yeast . Finally , lack of Swd2 . 2 and Sen1 resulted in a small but significant increase in the formation of R-Loops at some but not all Pol III-transcribed genes ( Figure S8 ) . Note however that this increase could be due to the fact that Pol III transcription is stimulated in the absence of Swd2 . 2 and Sen1 ( Figure 2C&D ) . As such , these observations therefore do not prove that Swd2 . 2 and Sen1 antagonize R-Loop formation at Pol III-transcribed genes directly . To establish whether R-Loops at Pol III-transcribed genes could contribute to the accumulation of condensin , we prevented the formation of stable R-Loops by over-expressing RnhA . ChIP analysis showed that over-expression of RnhA did not reduce the amount of condensin recruited at Pol III-transcribed genes in swd2 . 2Δsen1Δ cells ( Figure 3E ) or in wild-type mitotic cells ( Figure S9 ) . These data concur to demonstrate that stable , long-lived R-Loops play little or no part in recruiting condensin . Note that over-expression of RnhA did not interfere either with the association of RNA Pol III ( Figure S10A ) or Sen1 ( Figure S10B ) . Because Xenopus condensin shows greater affinity in vitro for positively supercoiled DNA [21] , we speculated that the cue facilitating the accumulation of condensin at Pol III-transcribed genes in the absence of Swd2 . 2 and Sen1 could be local topological constraints . Consistent with an increase in topological stress in swd2 . 2Δsen1Δ cells , ChIP analysis detected strong accumulation of topoisomerase I ( Top1 ) at most loci ( Figure 4A ) , although the protein levels of Top1 remained unaffected ( Figure 4B ) . We also detected enhanced accumulation of topoisomerase II ( Top2 ) , mostly at Pol III-transcribed genes ( Figure 4C ) , when the protein levels of Top2 remained unaffected ( Figure 4D ) . Transcription-associated topological stress was recently shown to destabilize nucleosomes [19] . At some but not all Pol III-transcribed genes that we tested , we detected a significant reduction in the recruitment of histone H3 ( Figure 4E ) in swd2 . 2Δsen1Δ cells , which is consistent with the local depletion of nucleosomes . The concomitant accumulation of Top1 and Top2 and the depletion of nucleosomes suggest that topological stress is greater at Pol III-transcribed genes in swd2 . 2Δsen1Δ cells . We speculate that the increased transcription of Pol III-transcribed genes in swd2 . 2Δsen1Δ cells could contribute at least in part to this enhanced topological stress . As R-Loops unwind the DNA , it was possible that the abundance of R-Loops formed at Pol III-transcribed genes ( Figure 3 ) could contribute to this topological stress . To test this possibility , we monitored by ChIP the localization of Top2 upon over-expression of RnhA . Surprisingly , the localization of Top2 was not altered at Pol III-transcribed genes upon over-expression of RnhA , whilst it was reduced at the Pol I-transcribed 18S ( Figure S11 ) . This suggested that the impact of R-Loop formation on the surrounding chromatin depends on where in the genome R-Loops form . Based on these results , we envisaged two possible models to explain the increased localization of condensin at Pol III-transcribed genes in the absence of Swd2 . 2 and Sen1: either the accumulation of Top1 and/or Top2 helps to recruit and/or stabilize condensin , or topological stress facilitates the association of condensin at Pol III-transcribed genes . We previously identified the deletion of Top1 ( top1Δ ) as a suppressor of cut3-477 [22] , suggesting that the accumulation of Top1 that results from lack of Swd2 . 2 and Sen1 is unlikely to facilitate the association of condensin with chromatin . Figures 5A&B show that the triple deletion swd2 . 2Δsen1Δtop1Δ was a better suppressor of cut3-477 than the double deletion swd2 . 2Δsen1Δ . This genetic evidence suggested that failure to monitor topological stress in top1Δ cells might facilitate the association/function of condensin . In support of this , ChIP analysis showed that there was a small but significant increase in the association of condensin at most Pol III-transcribed genes in cells deleted for Swd2 . 2 , Sen1 and Top1 ( swd2 . 2Δsen1Δtop1Δ cells ) ( Figure 5C ) . Taken together , these data support the following model: the absence of Swd2 . 2 and Sen1 increases the transcriptional activity at Pol III-transcribed genes and this might contribute to enhance local topological constraints . These constraints , either directly or indirectly , contribute to recruit or maintain condensin at Pol III-transcribed genes ( Figure 5D ) . To establish whether topological stress was sufficient to stimulate the association of condensin with chromatin , we monitored the association of condensin in the temperature-sensitive Top2 mutant top2-191 ( [33] ) at the semi-restrictive temperature of 28°C . This analysis showed that the association of condensin was not significantly disrupted in these conditions ( Figure S12A ) . Similarly , lack of Top1 on its own did not significantly impact the association of condensin ( Figure S12B ) . Taken together , these observations suggest that topological stress on its own is not sufficient to stimulate the association of condensin with chromatin .
In order to explain that condensin localizes to highly expressed genes from pro- to eukaryotes , whatever the RNA polymerase involved , we first hypothesized that a transcription by-product could facilitate the association of condensin with chromatin ( see Introduction ) . We speculated that this mechanism could represent the ancestral way of recruiting condensin to chromatin . Complementary cis-acting factors would then have evolved to stabilize the interaction of condensin with specific loci , as shown previously ( reviewed in [4] ) . In this study we specifically considered two transcription by-products as potential condensin-attracting features: R-Loop formation and transcription-associated topological stress . Both features have been described both in pro- and eukaryotes and they generate structures ( single-stranded DNA and positive supercoiling ) for which condensin has been shown to display high affinity in vitro . Our data are not consistent with the idea that stable R-Loops could be involved in recruiting condensin . Similarly , topological stress on its own was not sufficient to disrupt the localization pattern of condensin . However , our data show that topological stress facilitated the association of condensin at Pol III-transcribed genes when Swd2 . 2 and Sen1 were missing . These observations are consistent with the recent demonstration that supercoiling at highly expressed genes contributes to the establishment of topological domains and small-range chromosome compaction in Caulaboacter crescentus [34] . How could topological stress create a better binding site for condensin at Pol III-transcribed genes in the absence of Swd2 . 2 and Sen1 ? First , condensin might simply have a higher affinity for supercoiled chromatin , as suggested by the observation that condensin associates preferentially in vitro with positively supercoiled plasmids [21] . Alternatively , or in addition , topological stress might work by facilitating nucleosome eviction [19] . Consistent with the latter , budding yeast condensin associates preferentially with nucleosome-free regions , especially at Pol III-transcribed genes [12] . To explain that lack of Top1 only facilitates the association of condensin at Pol III-transcribed genes when Swd2 . 2 and Sen1 are missing , we speculate that the level of topological stress has to go over a certain threshold in order to attract/stabilize condensin . This threshold would be reached in the chromatin around Pol III-transcribed genes when Swd2 . 2 and Sen1 are missing but not when Top1 only is missing . The biology of R-Loops is a rapidly expanding field of investigation , and many observations now demonstrate that R-Loops control genome stability and gene expression in multiple ways ( reviewed in [35] ) . It is therefore essential to establish reliable methods to map R-Loop forming regions in genetically tractable organisms such as yeast to address the many functions of R-Loops in vivo . We presented evidence that the commonly used S9 . 6 ChIP method to map R-Loop forming regions in yeast is challenged by the fact that R-Loops , or at least their recognition by the S9 . 6 antibody , are partly sensitive to formaldehyde cross-linking and sonication . To circumvent this problem , we have developed two reliable alternatives to map R-Loop forming regions in fission yeast . Both of our methods concur to demonstrate that RNA-Pol III transcribed genes are major R-Loop forming regions in fission yeast . R-Loops have also been detected at Pol III-transcribed genes in budding yeast ( [28] ) , suggesting that R-Loop formation is a conserved feature of Pol III transcription , at least in yeast . We would like to argue that the two methods we have set up are complementary: not only do they map R-Loop forming regions but their use in parallel can also give information regarding the stability of R-Loops formed at different loci . Our data show that RNase H1 is most abundant at Pol III-transcribed genes throughout the cell-cycle , suggesting that R-Loops are constantly formed and detected by RNase H1 there . Our data also show that over-expression of RnhA in vivo counter-acts R-Loop formation more efficiently at Pol III-transcribed genes than within the rDNA for example ( 18S , Figure 3B ) . On the contrary , DRIP only yields significant signals at Pol III-transcribed genes when RNase H1 and RNase H2 are missing ( rnh1Δrnh201Δ cells ) , whilst the DRIP signals at the rDNA ( 18S ) are significant in wild-type cells , when RNase H1 and RNase H2 are fully active . At Pol III-transcribed genes , DRIP signals increase 10-20 fold in rnh1Δrnh201Δ cells , whilst they only increase ∼3-fold at the rDNA ( 18S ) . Our interpretation of these data is that R-Loops formed at 18S are stable and a relatively poor substrate for RNase H1 , whilst R-Loops formed at Pol III-transcribed genes are unstable and a good substrate for RNase H1 . A corollary to these observations is that DRIP is probably better suited to detect long-lived , stable R-Loops . This might explain why DRIP did not detect significant R-Loop formation at Pol III-transcribed genes in human cells ( [16] , [36] ) . We conclude that using both R-Loop mapping methods in parallel could provide indications of the relative stability of R-Loops at different loci . The reasons why R-Loops formed at Pol III-transcribed genes are labile are still unclear but we speculate that R-Loops formed at Pol III-transcribed genes might be smaller than those formed at the 18S because the Pol III transcription units are much smaller . Further studies will be required to understand the consequences of R-Loop formation at Pol III-transcribed genes and how the half-life of an R-Loop might influence its function . R-Loop formation has been shown to be associated with increased phosphorylation of histone H3 on Serine 10 and reduced chromatin accessibility [14] . In turn , the phosphorylation of histone H3 on Serine 10 facilitates the interaction between adjacent nucleosomes , thereby promoting chromatin compaction [37] . We showed previously that to constitutively increase the levels of histone H3 phosphorylated on Serine 10 by deleting PP1 phosphatase ( dis2Δ ) was not sufficient to significantly improve chromosome segregation when condensin was deficient [22] , suggesting that H3-S10-mediated chromatin compaction cannot compensate for the deficiency of condensin . Here we presented evidence that stable R-Loops do not significantly contribute to the recruitment of condensin . Taken together , these observations concur to establish that R-Loop-mediated chromatin compaction is distinct from condensin-mediated chromosome condensation . Our data also suggest that the action of condensin is more fundamental to building a mitotic chromosome than R-Loop-mediated chromatin compaction . Our data have highlighted unexpected ways by which proteins involved in the metabolism of RNA can affect chromosome segregation and genome integrity . Published data demonstrated conclusively that mutations in such factors in general and in Sen1 in particular resulted in chromosome instability ( CIN ) in yeast , in a mechanism involving R-Loop formation antagonizing replication fork progression ( [38] , [39] and reviewed in [35] ) . Here on the contrary , our data show that deletions of two such factors , Swd2 . 2 and Sen1 , facilitate the segregation and stability of chromosomes when condensin is deficient , in a mechanism that does not require stable R-Loop formation . In addition , our data show that Swd2 . 2 and Sen1 keep topological stress under control at Pol III-transcribed genes . We speculate that the enhanced transcription at Pol III-transcription associated with lack of Swd2 . 2 and Sen1 could contribute to such stress . However , we cannot exclude the possibility that RNA Pol III-dependent transcription is also defective in other ways that could explain the accumulation of topological stress when Swd2 . 2 and Sen1 are missing . The answer to this question will require further studies . Beautiful in vitro approaches demonstrated unequivocally that budding yeast Sen1 contributes to transcription termination of some RNA Pol II transcripts ( [26] ) . It is not yet known whether fission yeast Sen1 has the same function . As fission yeast Sen1 is not essential for viability whilst its budding yeast counterpart is , it is possible that the function of Sen1 has diverged in fission yeast . This idea is supported by our data showing that RNA Pol III is likely to be the most stable binding partner of Sen1 in fission yeast and that Sen1 antagonizes Pol III-dependent transcription . On the contrary , a recent study aimed at identifying the binding partners of RNA Pol III in budding yeast did not identify Sen1 , suggesting that the interaction between Sen1 and RNA Pol III is not as stable and/or abundant in budding yeast [40] . Further work is required to understand the function of fission yeast Sen1 at Pol III-transcribed genes . Previous studies had concluded that the inhibition of RNA Pol I or RNA Pol II in mitosis was a pre-requisite for the binding of condensin at repetitive sequences [9] , [10] , suggesting that a processive RNA polymerase is a hindrance to the binding of condensin on chromatin . Here we challenge this idea by showing that an enhanced recruitment of condensin at Pol III-transcribed genes is associated with an increase in the expression of the same genes . These data show that , at least at Pol III-transcribed genes , an active polymerase is not an obstacle for the binding of condensin .
A complete list of all of the strains used in this study is given in Table S3 . Standard genetic crosses were employed to construct all strains . Rnh1-GFP , Sen1-GFP , and Top1-3flag were generated using a standard PCR procedure . To obtain Rnh1D129N , Rnh1 was PCR amplified and cloned into pCRII ( Life technologies ) . Site-directed mutagenesis was then used to mutate the residue D129 into N ( GAC to AAC ) using Quickchange protocols ( Stratagene ) . Overlapping PCR was used to add a C-terminus GFP tag and a cassette of resistance to kanamycin ( KanR ) to the mutagenized Rnh1 in order to integrate the mutagenized Rnh1 at the endogenous Rnh1 locus . After yeast transformation , proper integrants were selected by PCR and western blot and were sequenced to verify the presence of the mutation . The plasmid over-expressing RnhA tagged with 1xFLAG at its N-terminus was obtained from Eun Shik Choi and Robin Allshire ( WTCCB , Edinburgh , UK ) . In order to stably integrate the plasmid in the genome , it was linearized by digestion with MluI and then transformed in to yeast according to standard procedures . 1 , 5 . 108 cells were treated with 1% formaldehyde ( Sigma ) at 17°C for 30′ . After extensive washes with cold PBS , cells were frozen in liquid Nitrogen . Frozen cells were then broken open using a RETSCH MM400 Mill and then resuspended in cold lysis buffer ( Hepes-KOH 50 mM pH 7 , 5 , NaCl 140 mM , EDTA 1 mM , Triton 1% , Na-deoxycholate 0 , 1% , PMSF 1 mM ) . The lysats were then sonicated at 4°C using a Diagenode sonicator . Immuno-precipitation was done overnight at 4°C using Protein A-coupled Dynabeads previously incubated with the anti-GFP A11122 antibody ( Invitrogen ) or using Protein G-coupled Dynabeads previously incubated with the anti-myc 9E10 antibody ( Sigma ) according to the manufacturer's instructions . Beads were washed successively with ( 5′ incubation on rotating wheel ) : Wash I buffer ( 20 mM Tris pH 8 , 150 mM NaCl , 2 mM EDTA , 1% Triton-X100 , 0 , 1% SDS ) , Wash II buffer ( 20 mM Tris pH 8 , 500 mM NaCl , 2 mM EDTA , 1% Triton-X100 , 0 , 1% SDS ) and Wash III buffer ( 20 mM Tris pH 8 , 1 mM EDTA , 0 , 5% Na-deoxycholate , 1% Igepal , 250 mM LiCl ) . After two additional washes in TE pH 8 , the beads were resuspended in 10% Chelex resin ( Biorad ) and incubated at 98°C for 10′ . After addition of 2 µL of 10 mg/mL of proteinase K , the mixture was incubated at 43°C for 1 hour , then for another 10 mn at 98°C . After centrifugation , the supernatant was collected and analyzed by qPCR . 8 . 108 cells were frozen in liquid nitrogen , broken open using a RETSCH MM400 Mill and then resuspended in cold lysis buffer ( Hepes-KOH 50 mM pH 7 , 5 , NaCl 140 mM , EDTA 1 mM , Triton 1% , Na-deoxycholate 0 , 1% ) . After phenol/chloroform purification and ethanol precipitation , the DNA was resuspended in TE pH 8 and split into two samples . Both samples were digested with BsrGI , EcoRI , HindIII , SspI and XbaI according to the manufacturer's instructions and RNase H was added to one of the two samples . After digestion , each sample was divided into two and incubated overnight at 4°C in IP buffer ( 100 mM MES pH 6 , 6 , NaCl 500 mM , 0 , 05% Triton , 2 mg/mL BSA ) in the presence of either Protein A-coupled Dynabeads or Protein A-coupled Dynabeads previously incubated with the S9 . 6 antibody according to the manufacturer's instructions . The beads were then washed three times in IP buffer . After two additional washes in TE pH 8 , the beads were resuspended in 10% Chelex resin ( Biorad ) and incubated at 98°C for 5′ . After addition of 2 µL of 10 mg/mL of proteinase K , the mixture was incubated at 43°C for 30′ , then for another 5′ at 98°C . After centrifugation , the supernatant was collected and analyzed by qPCR . Immunoprecipitation was carried out as described previously [22] , except that cells were broken open using a RETSCH MM400 Mill . To purify Sen1-associated proteins ( Table S2 ) , a protein extract was prepared from 109 cells expressing GFP-tagged Sen1 from the endogenous locus . After immuno-precipitation with 15 µL of magnetic beads , the beads were washed three times with 1 mL of lysis buffer and twice with 1 mL of PBS containing 0 , 02% Tween . The beads samples were then subjected to in-solution reduction , carbamidomethylation and tryptic digestion . After acidification with 10%Trifluoroacetic Acid the samples were centrifuged 3 times to eliminate the beads . Peptide sequences were determined by mass spectrometry performed using a LTQ Velos instrument ( Dual Pressure Linear Ion Trap ) equipped with a nanospray source ( Thermo Fisher Scientific ) and coupled to a U3000 nanoLC system ( Thermo Fisher Scientific ) . A MS survey scan was acquired over the m/z range 400–1600 in Enhanced resolution mode . The MS/MS scans were acquired in Normal resolution mode over the m/z range 65–2000 for the 20 most intense MS ions with a charge of 2 or more and with a collision energy set to 35eV . The spectra were recorded using dynamic exclusion of previously analyzed ions for 0 . 5 min with 50 millimass units ( mmu ) of mass tolerance . The peptide separation was obtained on a C18 PepMap micro-precolumn ( 5 µm; 100 Å; 300 µm×5 mm; Dionex ) and a C18 PepMap nanocolumn ( 3 µm; 100 Å; 75 µm×200 mm; Dionex ) using a linear 90 min gradient from 0 to 40% , where solvent A was 0 . 1% HCOOH in H2O/CH3CN ( 95/5 ) and solvent B was 0 . 1% HCOOH in H2O/CH3CN ( 20/80 ) at 300 nL/min flow rate . Protein identification was performed using the MASCOT Algorithm from the Proteome Discoverer software v1 . 1 ( Thermo Fisher Scientific ) against the UniProtKB database reduced to Schizosaccharomyces pombe species [UniProt release 2013_12] . These were performed as previously described [22] . | Failure to condense chromosomes prior to anaphase onset can lead to genome instability . The evolutionary-conserved condensin complex drives chromosome condensation , probably by changing the topology of chromatin around its binding sites . Condensin localizes to regions of high transcription , suggesting that some transcription-associated feature ( s ) direct its association with chromatin . Here we considered that transcription-dependent DNA:RNA hybrids or topological stress could be involved in recruiting condensin . Our data show that condensin is indeed enriched at regions accumulating DNA:RNA hybrids but that they are not involved in its recruitment . Rather , we identify a mutant combination where increased transcription by RNA Pol III is associated locally with stronger topological stress . Strikingly the localization of condensin is dramatically enhanced at the same loci and we show that topological stress contributes to this enhanced association . Our data strengthen the idea that transcription creates the environment necessary to recruit condensin in mitosis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"chromosome",
"structure",
"and",
"function",
"gene",
"regulation",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"dna",
"transcription",
"mitosis",
"fungi",
"chromatin",
"schizosaccharomyces",
"chromosome",
"biology",
"gene",
"expression",
"schizosaccharomyces",
"pombe",
"molecular",
"biology",
"yeast",
"biochemistry",
"rna",
"rna",
"processing",
"cell",
"biology",
"transcriptional",
"termination",
"genetics",
"biology",
"and",
"life",
"sciences",
"organisms",
"chromosomes"
] | 2014 | RNA Processing Factors Swd2.2 and Sen1 Antagonize RNA Pol III-Dependent Transcription and the Localization of Condensin at Pol III Genes |
Cleft palate ( CP ) is one of the most commonly occurring craniofacial birth defects in humans . In order to study cleft palate in a naturally occurring model system , we utilized the Nova Scotia Duck Tolling Retriever ( NSDTR ) dog breed . Micro-computed tomography analysis of CP NSDTR craniofacial structures revealed that these dogs exhibit defects similar to those observed in a recognizable subgroup of humans with CP: Pierre Robin Sequence ( PRS ) . We refer to this phenotype in NSDTRs as CP1 . Individuals with PRS have a triad of birth defects: shortened mandible , posteriorly placed tongue , and cleft palate . A genome-wide association study in 14 CP NSDTRs and 72 unaffected NSDTRs identified a significantly associated region on canine chromosome 14 ( 24 . 2 Mb–29 . 3 Mb; praw = 4 . 64×10−15 ) . Sequencing of two regional candidate homeobox genes in NSDTRs , distal-less homeobox 5 ( DLX5 ) and distal-less homeobox 6 ( DLX6 ) , identified a 2 . 1 kb LINE-1 insertion within DLX6 in CP1 NSDTRs . The LINE-1 insertion is predicted to insert a premature stop codon within the homeodomain of DLX6 . This prompted the sequencing of DLX5 and DLX6 in a human cohort with CP , where a missense mutation within the highly conserved DLX5 homeobox of a patient with PRS was identified . This suggests the involvement of DLX5 in the development of PRS . These results demonstrate the power of the canine animal model as a genetically tractable approach to understanding naturally occurring craniofacial birth defects in humans .
Cleft palate ( CP ) is one of the most commonly occurring craniofacial birth defects , affecting approximately 1 in 1500 live human births in the United States [1] . Children born with CP may develop hearing loss , difficulties with speech and eating , and may be at an increased risk for psychiatric disorders and neurological deficits [2]–[4] . CP occurs when there is a failure in the formation of the secondary palate , which makes up all of the soft palate and majority of the hard palate . Secondary palate development is conserved across mammalian species and proceeds through highly regulated sequential steps: palatal shelf growth , elevation , fusion , and cell differentiation ( reviewed in [5] ) . Disruptions in any of these pathways may cause a cleft palate and lead to the phenotypic spectrum of CP cases that is observed . CP may occur alone ( nonsyndromic ) or with other abnormalities ( syndromic ) . Pierre Robin sequence ( PRS , OMIM 261800 ) is a heterogeneous and phenotypically variable subgroup of CP that affects 1 in 8500 live human births [6] . PRS is characterized by a cleft palate , shortened mandible ( micrognathia ) , and posteriorly placed tongue ( glossoptosis ) . PRS is thought to be the result of a sequence of events caused by the primary defect , micrognathia [7] . The etiology of PRS is still largely unknown and highly variable . PRS may occur alone or as part of a syndrome , such as Stickler syndrome , Velocardiofacial syndrome , and Treacher Collins syndrome [8] , [9] . A high incidence within families and among twins suggests a genetic etiology . Familial aggregation with an autosomal dominant mode of inheritance has been observed with translocations of 17q24 and a reduction of SOX9 and KCNJ2 expression [10] , [11] . PRS also occurs at a high incidence among families with cleft lip and palate [12] , [13] . However , monozygotic twins discordant for PRS are also observed , suggesting that PRS may be a result of the twinning process or of mandibular constraint in utero [13] , [14] . In an effort to understand the genetic basis of craniofacial birth defects such as PRS , we used a relatively unconventional model organism , the domestic dog . Dogs have naturally occurring birth defects , with inherited orofacial clefts that resemble those observed in humans [15]–[17] . Domestication and subsequent pedigreed breed creation from a small number of founders has led to a unique genetic background , resulting in a small number of genetic variants being responsible for the broad phenotypic diversity observed [18] . Compared to humans , dogs are amendable to association-based mapping studies with a small number of samples due to their relatively long linkage disequilibrium blocks within breeds [19] . Here , we demonstrate how a naturally occurring model of PRS in the Nova Scotia Duck Tolling Retriever ( NSDTR ) breed , characterized by CP and relative micrognathia , led to the identification of both the first mutation responsible for cleft palate in dogs and candidate genes for PRS in people .
DNA samples were collected from 14 NSDTRs that had clefts of the hard and soft palate . To identify loci associated with the CP phenotype in the NSDTR , a genome-wide association was performed in 14 CP NSDTR cases and 72 controls across ∼173 , 000 SNPs . After quality control , 109 , 506 SNPs remained . Chi-square analysis of the remaining SNPs identified an associated region on canine chromosome 14 ( cfa14; Figures 1A and 1B ) . A homozygous 5 . 1 Mb haplotype was identified ( 24 . 2 Mb to 29 . 3 Mb ) in 12 of the 14 CP NSDTR cases ( Figure 1C ) . This homozygous haplotype is absent in all 72 controls . Parents ( n = 5 ) and littermates ( n = 6 ) of the 12 CP NSDTRs were heterozygous for the associated haplotype , suggesting a recessive mode of inheritance . Quantile-quantile ( Q-Q ) plots and a genomic inflation factor of 1 . 05 indicate that there is no underlying population stratification ( Figure 1D ) . To confirm that only the association is responsible for the deviation from the line of the null hypothesis , SNPs on cfa14 were removed and the Q-Q plot regenerated . There is little evidence of population stratification with a recalculated genomic inflation factor of 1 . 02 . Using microsatellite markers from the cfa14 interval , linkage analysis was performed on a subset of CP NSDTRs with the associated haplotype and available family members ( n = 8 ) . LOD scores were calculated under a fully penetrant recessive mode of inheritance ( Table S1 ) . A significant LOD score of 3 . 18 was obtained with a recombination fraction ( Θ ) of zero at cfa14 . 25006375 , further confirming the association . All 14 CP NSDTRs had clefts of the hard and soft palate ( Figures 2A and 2B ) , but the 12 CP NSDTRs with the associated haplotype exhibited a specific phenotype , which we designate CP1 ( Cleft Palate 1 ) . Micro-computed tomographic ( micro-CT ) imaging was performed to investigate additional craniofacial abnormalities on available skulls from 4 neonatal CP1 NSDTRs and 3 neonatal normal NSDTRs with the homozygous wildtype haplotype . One of the mandibles from the 4 CP1 NSDTRs was unavailable for imaging . Uneven alignment of the upper and lower jaw was noted in the 3 CP1 NSDTRs when compared to the 3 normal NSDTRs indicating relative mandibular brachygnathia ( Figures 2C and 2D ) . 3D measurements of the mandible length indicated that the CP1 NSDTRs had relatively shorter mandibles by an average of 5 . 46 mm when compared to the normal NSDTRs ( Table S2 ) . In the 4 cases , clefts were characterized by abnormal or missing palatine fissures , missing or small palatine processes of the maxilla , and small , missing , or abnormally shaped palatine bones ( Figures 2E–2G ) . The nasal septum was absent or poorly developed . In the three CP1 NSDTR with mandibles , variation from the normal angulation of the condylar process was observed ( Figure 2H ) . In two of these cases , an abnormal angle of the mandibular head of the condylar process was observed . One case had an additional general asymmetry of the entire craniofacial complex . Based on the phenotypic findings , we hypothesize that the 12 CP1 NSDTRs are animal models for cleft palate and micrognathia disorders . Although unavailable for micro-CT imaging , the 2 CP cases without the associated haplotype exhibited a normal jaw relationship with no relative mandibular brachygnathism . Since these dogs are phenotypically different , they were excluded from the rest of the analysis . Located within the interval defined by the genome-wide association study are 21 genes ( Table S3 ) . DLX5 and DLX6 ( cfa14: 25014704-25033706 ) were selected for sequencing due to their roles as transcription factors in craniofacial development and the similar phenotypes observed in mutant mice [20]–[22] . The coding regions and intronic regions with high conservation across species of DLX5 and DLX6 were sequenced in one CP1 case and one unaffected control NSDTR . One intronic nucleotide insertion ( 25032667-25032667insG ) was identified within DLX5 in both the CP1 case and unaffected control NSDTRs when compared to the CanFam 2 . 0 Boxer reference sequence [19] . Upon further investigation in additional NSDTRs , it was observed that this insertion does not segregate with the phenotype and is likely breed specific . A 2056 bp insertion was identified within a highly conserved region of DLX6 intron 2 at cfa14 . 25016716 in the CP1 case when compared to the unaffected control and reference sequence . The insertion is 82 bp into intron 2 and is flanked by a 13 bp target site duplication . A BLAST search identified 99% similarity with a LINE-1 element ( GenBank AC187025 . 7 ) [23] . Transcript analysis was performed in cDNA from cerebral cortex for both DLX5 and DLX6 . DLX5 and DLX6 transcripts were Sanger sequenced in one neonatal CP1 NSDTR with the LINE-1 insertion and compared to one neonatal unaffected control NSDTR . No polymorphisms were identified within DLX5 . cDNA from the CP1 NSDTR showed both the wildtype DLX6 transcript and a larger transcript , which contained 1281 base pairs of the intronic LINE-1 insertion ( Figure 3B ) . In order to quantify the amount of both wildtype DLX6 transcript and the larger mutant DLX6 transcript , real-time PCR of DLX6 from cerebral cortex cDNA was performed in 3 neonatal CP1 NSDTRs and 3 neonatal unaffected control NSDTRs . REST analysis indicated that in CP1 NSDTRs , when compared to control NSDTRs , the DLX6 wildtype transcript was downregulated by a mean factor of 0 . 252 ( p = 0 . 069 ) , while the DLX6 mutant transcript was significantly upregulated by a mean factor of 60 . 033 ( p = 0 . 012; Figures 3C and 3D ) [24] . Real-time PCR was also performed on DLX5 in the same samples . There was no significant change in DLX5 expression levels between CP1 cases and unaffected controls in the tissues examined . Sequence analysis and translation of the approximately 1 . 2 kilobase LINE-1 insertion predicts an in frame stop codon after the insertion of a new exon ( Figure 4 ) . A premature stop codon is predicted to truncate 17 amino acids from the 60 amino acid functional DNA binding homeodomain . In order to investigate segregation of the DLX6 LINE-1 insertion , PCR genotyping was performed on available DNA from parents , littermates , and the 12 CP1 NSDTRs ( Figure 5 ) . All 12 CP1 NSDTRs with the associated haplotype were homozygous for the LINE-1 insertion or mutant allele . Within families of the 12 CP1 NSDTRs , nine parents and 17 littermates were available for genotyping . Nine parents and 14 littermates were heterozygous for the mutant allele , while the 3 remaining littermates were homozygous for the wildtype allele . This indicates that the DLX6 LINE-1 insertion segregates both with the phenotype and with an autosomal recessive mode of inheritance . Additional dogs were PCR genotyped to determine the allele frequency of the DLX6 LINE-1 insertion ( Table 1 ) . Within the NSDTR breed , 96 dogs were genotyped and 80 NSDTRs did not carry the insertion , while the remaining 16 NSDTRs were heterozygous for the insertion . To determine if the insertion was shared among other breeds , 35 affected dogs from 20 other breeds and 284 unaffected dogs from 69 breeds were genotyped . No carriers were identified . This is consistent with a fully penetrant autosomal recessive causative mutation that is private to the NSDTR breed . To determine if variation in DLX5 or DLX6 contributed to cleft palate in humans , a cohort of patients with a variety of manifestations of cleft palate were sequenced . Sequencing of 59 patients with nonsyndromic CP ( NSCP ) and 92 patients with nonsyndromic clefting of the lip and palate ( NSCLP ) identified 7 novel intronic or 3′UTR variants ( Table 2 ) . The intronic variant at chr9: 96635849 was present in 1 . 25% of chromosomes and was absent from 1000 Genomes ( p = 0 . 03 ) [25] . However , none of the variants was significantly associated when correcting for multiple comparisons . To test for overtransmission of a common allele to affected offspring , a set of 362 case-parent trios from the US were genotyped across a subset of SNPs ( rs2272280 , rs3801290 , and rs3213654 ) and tested with TDT analysis . These SNPs were not significantly associated with NSCLP or NSCP ( p>0 . 4 , data not shown ) . The screening of 31 patients with syndromic CP identified 4 novel rare variants , two of which were protein coding ( Table 2 ) . These include a synonymous variant at p . Phe282 and a novel missense variant in a patient with PRS ( DLX5: p . Ile192Met; NM_005221 . 5:c . 576C>G ) . The missense variant was predicted to be damaging by SIFT , but was inherited from the affected individual's unaffected mother . However , this variant affects a highly conserved residue of the DNA-binding domain of DLX5 that is conserved across vertebrates and was absent from both the 1000 Genomes and NHLBI ESP databases [25] , [26] . An additional 15 patients with Pierre Robin sequence were sequenced , but no additional mutations were identified .
A naturally occurring animal model for PRS was discovered in twelve cases of CP1 NSDTRs , which exhibit relative mandibular brachygnathia and cleft palate . A genome-wide association study within NSDTRs with CP identified a shared 5 . 1 Mb homozygous haplotype among 12 CP1 cases with relative mandibular brachygnathia . From the associated interval , DLX5 and DLX6 were selected as regional candidate genes based on their roles in development . Sequencing of these regional candidate genes in NSDTRs identified an intronic LINE-1 insertion within DLX6 that segregates with the phenotype in the breed . This discovery prompted the sequencing of the same genes in a human cohort of CP cases , which identified a damaging missense mutation within the homeobox of DLX5 in a patient with PRS was identified . The regional candidate genes , DLX5 and DLX6 , make up a pair of convergently transcribed homeobox containing transcription factors [27] . They are involved in patterning of craniofacial structures , inner ear , limb , and brain development with roles in chondrocyte and osteoblast differentiation [21] , [28] , [29] . The functions of Dlx5 and Dlx6 were studied by targeted inactivation of murine homologs . Developmental expression of both genes was observed in the first pharyngeal arch , brain , and skeleton [27] . Single gene mutants ( Dlx5−/− and Dlx6−/− ) were perinatal lethal and resulted in brain , bone , and inner ear defects , with craniofacial abnormalities including a cleft palate , hypoplastic condylar process , and shortened mandible [20]–[22] . Double mutants ( Dlx5−/−;DLX6−/− ) exhibited more severe craniofacial , inner ear , and bone defects , and were a phenocopy of split hand/foot malformation [28] . They also exhibited homeotic transformation of the mandible into a maxilla indicating that these genes were responsible for normal patterning of the mandible [21] , [28] , [29] . Phenotypic similarity between CP1 NSDTRs , the PRS patient , and mutant mice are observed with the changes in the condylar process , cleft palate , and relatively shortened mandibles . There may be associated condylar hypoplasia in PRS , but this was not investigated in the PRS patient [30] , [31] . Shortened mandibles have been observed to cause cleft palate in mouse studies [7] , [32]–[34] . Palatal development is a highly regulated sequence of events involving the repositioning of the palatal shelves from a lateral to horizontal orientation over the tongue . During this process , the tongue repositions by elongation of the mandible , allowing for reorientation of the palatal shelves ( reviewed in [5] ) . If the mandible does not elongate , the tongue cannot reposition , which in turn obstructs palatal shelf elevation and leads to a cleft palate [7] . This supports the role of DLX5 and DLX6 in the phenotype observed in the PRS patient and CP1 NSDTRs . This also suggests that the CP1 NSDTRs are naturally occurring animal models for PRS because , although other disorders may be associated with cleft palate and micrognathia , these dogs lack additional abnormalities [35] . DLX5 and DLX6 contain homeoboxes that regulate transcription by binding to specific DNA sequences . Homeoboxes are highly conserved nucleotide sequences of 180 base pairs . Point mutations located within the homeobox give rise to disease at a higher frequency than mutations located within the rest of the gene and often show a dominant effect due to haploinsufficiency [36] . DLX5 is sensitive to haploinsufficiency because , although Dlx5+/− mice appear normal , closer investigation indicates that they have a decreased bone mineral density [20] , [37] . Mutations within the homeobox may also affect proper structural folding of the protein and DNA binding specificity leading to a mutant phenotype [38] . This is observed in a homozygous missense mutation in the homeobox of DLX5 in two affected family members in a consanguineous pedigree with split-hand/foot malformation ( SHFM ) [39] . The affected individuals were noted to have clefts of the hands , but no cleft palate or craniofacial abnormalities with the exception of a mildly pronounced forehead . The heterozygous missense mutation identified in the PRS patient ( NM_005221 . 5:c . 576C>G ) is located within the homeobox sequence of DLX5 , which supports it contribution to the observed phenotype . The DLX5 heterozygous missense mutation within the PRS patient does not segregate with a Mendelian mode of inheritance , but this is not expected since PRS is a complex trait . PRS was not apparent in the mother , but she may have exhibited subtle craniofacial abnormalities that went unnoticed . This likely complex inheritance suggests a role for additional , not yet identified loci or environmental factors . Support for this is observed when mice with heterozygous expression of a Dlx5 null allele and targeted inactivation of a transcription factor responsible for clefting in people , Msx1 , exhibit cleft palate ( Dlx5−/+; Msx1−/− ) [40] . However , when expression of both genes is disrupted ( Dlx5−/−; Msx1−/− ) , mice have no cleft palate . Nine noncoding sequence variants were identified within DLX5 and DLX6 of human patients with orofacial clefts . Functional analysis of the nine variants may provide further insight into the possible contribution of these genes to a cleft palate phenotype . Noncoding regions have important regulatory functions as they have been observed to disrupt gene expression . Reduced DLX5 and DLX6 expression was observed in an autistic patient with a SNP in the DLX5 and DLX6 intergenic region [41] . Hearing loss and craniofacial defects including cleft palate and micrognathia have been observed in the deletion of a DLX5 and DLX6 enhancer region [42] . The LINE-1 insertion within DLX6 of the CP1 NSDTRs is also located within a noncoding region that is highly conserved . According to the UCSC genome browser , the HMR conserved transcription factor binding site regulation track indicates that the LINE-1 insertion identified within DLX6 CP1 NSDTRs disrupts a binding domain for SUZ12 . SUZ12 is a long noncoding RNA that encodes a core of the polycomb repressive complex2 that has regulatory functions in the developing embryo [43] . The exact interaction of SUZ12 and DLX6 is not yet known , indicating that the LINE insertion may do more than disrupt transcription . LINE elements are transposable elements that comprise 21% of the human genome , 16% of the dog genome , and often insert into intronic regions [44] , [45] . Intronic LINE element insertions are observed to cause disease in Duchenne-like muscular dystrophy and chronic granulomatous disease through the introduction of a new exon [46]–[48] . cDNA sequence from CP1 NSDTRs homozygous for the LINE-1 insertion indicate that 1 kb of the LINE element is spliced into the DLX6 transcript . Premature protein truncation of DLX6 is predicted due to an in frame stop codon after the formation of a new exon . Both the Dlx6 mutant mice ( Dlx6−/− ) and CP1 NSDTRs phenotypes are the result of truncating the same 3′ sequence of the homeodomain [22] . The LINE insertion disrupts transcription of DLX6 within CP1 NSDTRs and leads to downregulation of wildtype DLX6 transcript when compared to unaffected NSDTRs . As a result , , only 25% of the normal expression levels are produced . The reduced expression of wildtype DLX6 transcript is not enough to prevent CP and the mandibular abnormalities . DLX5 and DLX6 expression levels have been observed to vary based on the timepoint and tissue examined with complex transcriptional regulation from tissue specific enhancers and noncoding RNAs [20] , [21] , [27] , [28] , [42] , [49]–[51] . It is possible that the LINE insertion works to disrupt appropriate timing of tissue specific expression since it is inserted into a highly conserved region within intron 2 . Although not statistically significant , expression analysis from additional biological replicates from the correct tissue during the correct embryonic timepoint would likely yield significant values in the CP1 NSDTRs . This discovery provides a genetic tool for the NSDTR breeder who wishes to avoid producing cleft palate affected puppies . The LINE insertion identified within the CP1 NSDTRs is unique to a subset of cases with CP within the breed and cannot be used as a tool to prevent against all genetic causes of CP . CP disease heterogeneity is observed in the NSDTR breed as 2 of the 14 CP cases did not possess relatively shortened mandibles and the associated LINE-1 insertion . This is unexpected in a purebred dog breed with few founders where the inbreeding coefficient is 0 . 26 [52] . This indicates that even in relatively genetically homogenous samples , heterogeneity and a complex etiology that mimics human cleft cases such as PRS is observed . This highlights the promise of the dog as an animal model for birth defects to identify multiple genes and/or pathways involved in craniofacial development . In summary , identifying a mutation in an animal model with naturally occurring birth defects has enabled the identification of new candidate genes for PRS in people . This supports the continued use of the naturally occurring birth defects found within dogs and their unique genetic background to further our understanding of commonly occurring birth defects in people .
The use of samples involving human participants was approved by the institutional review board at the University of Iowa ( approval #s 199804080 and 199804081 ) . Informed consent was obtained and all clinical investigation was conducted according to the principles expressed in the Declaration of Helsinki . The collection of canine samples used in this study was approved by the University of California , Davis Animal Care and Use Committee ( protocol #16892 ) . Blood and tissue samples from NSDTRs with cleft palate ( n = 14 ) , healthy littermates ( n = 24 ) , parents ( n = 11 ) , unaffected NSDTRs ( n = 153 ) , and dogs with cleft palate across 20 breeds ( n = 35 ) were collected from privately owned animals . When available , tissue was collected at post mortem examination and flash frozen . The evaluations of orofacial clefts were performed by visual inspection of affected dogs . Blood samples from unaffected dogs ( n = 284 ) across 69 other breeds were collected from the William R . Pritchard Veterinary Medical Teaching Hospital . Genomic DNA was extracted from EDTA whole blood and tissue samples using Gentra Puregene DNA purification extraction kit ( Qiagen , Valencia , CA ) . Genome-wide SNP genotyping was performed with 14 cases and 72 controls using the Illumina CanineHD BeadChip ( Illumina , San Diego , CA ) with 173 , 662 markers . Samples had a genotyping call rate of ≥90% . 63 , 195 SNPs were excluded due to a minor allele frequency of ≤0 . 05 and 2 , 994 SNPs were excluded for a missing call rate of ≤10% . Chi-square analysis was performed using Plink [53] . 100 , 000 permutations were performed to correct for multiple tests . To determine if population stratification was present , quantile- quantile plots were generated in R using Plink output data [54] . The genomic inflation factor was also calculated using Plink . High-resolution micro-computed tomography ( micro-CT ) was used to evaluate craniofacial structures in 4 CP1 NSDTRs that were homozygous for the LINE-1 insertion , and in 3 normal NSDTRs homozygous for the wildtype allele . Samples were imaged at the Center for Molecular and Genomic Imaging ( UC Davis ) . The skulls were kept in zip-lock bags and allowed to warm to the CT scanner temperature ( 29°C ) inside of a custom plastic holder . CT images were obtained on the Centers MicroXCT-200 specimen CT scanner ( Carl Zeiss X-ray Microscopy ) . The CT scanner has a variable x-ray source capable of a voltage range of 20–90 kV with 1–8W of power . Samples were mounted on the scanners sample stage , which has submicron level of position adjustments . Scan parameters were adjusted based on the manufacturer's recommended guidelines: source and detector distances were 108 mm and 35 mm , respectively; the manufacturers LE4 custom filter was used for beam filtration; the voltage and power were set to 70 kV and 8W , respectively; and 1600 projections were acquired over 360-degrees with an exposure time of 1 . 5 s . Images were reconstructed on an isotropic voxel grid with 51 . 1507 microns per edge . Digital TIFF images were imported into Amira 5 . 4 . 5 ( Visualization Sciences Group , FEI ) For all specimens , tridimensional reconstructive ( 3D ) images were generated to assess the spatial relationship of the bones . 3D length measurements were performed using the 3D length tool . Visual inspection of the micro-CT images was performed to identify any abnormalities . Four regional microsatellites were mined from the UCSC genome browser ( CanFam 2 . 0 ) “Variation and Repeats” database within cfa14: 24 . 2 Mb–29 . 3 Mb . Fluorescently labeled microsatellite primers were designed ( Table S4 ) using Primer3 [55] . Microsatellites were PCR amplified using the following protocol: 94°C for 12 minutes , 7 cycles of 93°C for 20 seconds , 65°C for 30 seconds , 72°C for 2 minutes , 5 cycles of 93°C for 20 seconds , 58°C for 30 seconds , 72°C for 2 minutes , and 25 cycles of 93°C for 20 seconds , 55°C for 30 seconds , 72°C for 2 minutes , followed by a final annealing of 72°C for 2 minutes . Genotyping was performed on an ABI 3100 Genetic Analyzer ( Applied Biosystems , CA ) in 8 CP1 NSDTRs and 32 control NSDTRs . All genotypes were analyzed using STRand software [56] . LOD score analysis of microsatellite data was performed using Mendel software's LOCATION_SCORES option in 59 individuals from three pedigrees [57] . The disease trait was coded as autosomal recessive and assumed to be fully penetrant . Primers to amplify the exons , intron/exon boundaries , and regions of high conservation within the intragenic and intergenic regions of DLX5 and DLX6 were designed in Primer3 ( Table S3 ) [55] . PCR products were amplified in one CP1 NSDTR and one unaffected NSDTR . Areas with high GC content were amplified using Invitrogen AccuPrime GC-Rich DNA Polymerase protocols ( Life Technologies , Grand Island , NY ) . PCR products were cleaned using ExoSAP-IT and sequenced using the Big Dye terminator mix on an ABI 3100 Genetic Analyzer ( Applied Biosystems , CA ) . Sequences were analyzed using Chromas ( Technelysium , Tewantin , QLD , Australia ) and Vector NTI ( Informax , Frederick , MD ) . Sequences were aligned to each other and the Boxer reference sequence ( CanFam 2 . 0 ) to identify any polymorphisms [19] . The DLX6 LINE-1 insertion was amplified using LongAmp Taq PCR Kit ( New England BioLabs Ipswich , MA ) and cloned using the Invitrogen TOPO TA Cloning kit ( pCR2 . 1-TOPO vector ) with One Shot TOP10 Chemically Competent E . coli . Products were isolated with the Qiaprep Spin Miniprep kit ( Qiagen , Valencia , CA ) and sequenced using an ABI 3100 Genetic Analyzer ( Applied Biosystems , CA ) . LINE element sequence was identified using BLAST 2 . 2 . 28 [23] . Expression of DLX5 and DLX6 was evaluated in NSDTR lymphocytes , unaffected adult beagle tissue from cerebellum , cerebral cortex , heart , kidney , liver , skeletal muscle , skin , spinal cord , spleen , testis , and thymus . Total RNA was isolated from tissue samples using Qiagen QIAamp Blood Mini Kit ( Valencia , CA ) tissue protocols . Adult beagle total RNA was obtained from Zyagen ( San Diego , CA ) . RNA was synthesized into cDNA using Invitrogen Superscript III First Strand Synthesis System to RT PCR kit ( Life Technologies , Grand Island , NY ) . GAPDH was amplified ( forward primer 5′AAGATTGTCAGCAATGCCTCC 3′ and reverse primer 5′ CCAGGAAATGAGCTTGACAAA 3′ ) in these tissues to ensure that equivalent amounts of cDNA were added . Invitrogen 5′ prime RACE system for Rapid Amplification of cDNA ends kit ( Life Technologies , Grand Island , NY ) was used to sequence the 5′ prime end of DLX6 from embryo cDNA . RACE primers were designed using Primer3 ( Table S3 ) [55] . RACE PCR products were cloned using the Invitrogen TOPO TA Cloning kit ( pCR2 . 1-TOPO vector ) with One Shot TOP10 Chemically Competent E . coli . Products were isolated with the Qiaprep Spin Miniprep kit ( Qiagen , Valencia , CA ) and sequenced using an ABI 3100 Genetic Analyzer ( Applied Biosystems , CA ) . Genotyping primers were designed using Primer3 [55] . PCR genotyping was performed using a shared FAM labeled forward primer ( 5′ ACCATCGCTTTCAGCAAACT 3′ ) . Unlabeled reverse primers specific for the LINE-1 insertion ( 5′ GCAACTAATATTCGATAAAGCAGAA 3′ ) and wildtype ( 5′ CTAGGCCCAGAATTCCTCCT 3′ ) were designed . The PCR program was as follows: 94°C for 12 minutes , 35 cycles of 94°C for 30 seconds , 58°C for 30 seconds , and 72°C for 45 seconds , followed by 72°C for 20 minutes . The wildtype product produced a 171 base pair product and the mutant product produced a 184 base pair product . GeneScan 500 ROX size standards were used and the reaction was analyzed on an ABI 3100 Genetic Analyzer ( Applied Biosystems , CA ) . 96 unrelated unaffected NSDTRs and 284 unaffected dogs across 69 breeds were genotyped for the insertion . Nonsyndromic cleft palate cases ( n = 35 ) across 20 breeds were genotyped . All genotypes were analyzed using STRand software [56] . Primer sequences were generated using Primer3Plus ( http://primer3plus . com/ ) . A shared forward primer ( 5′aaactcagtacctggcccttc 3′ ) was designed with reverse primers unique to wildtype ( 5′ccatatcttcacctgtgtttgtg 3′ ) and mutant ( 5′aaactcagtacctggcccttc 3′ ) . Semi quantitative PCR using AmpliTaq Gold DNA Polymerase was performed to test the quality of cDNA and primers , to confirm product size and to check for the presence of genomic DNA contamination . Real-time PCR was performed using the Rotor-Gene SYBR Green PCR Kit ( QIAGEN , Valencia , CA ) using a 2-step cycle protocol ( 45 cycles; Initial denaturation-5 minutes at 95°C; Annealing- 15 seconds at 95°C; Extension- 90 seconds at 95°C; Final Melt curve ) on the Rotor Gene Q real-time PCR system . Cerebral cortex from 3 neonatal unaffected NSDTRs and 3 neonatal CP1 NSDTRs were run in triplicates with each replicate containing 0 . 2 ng template cDNA . cDNA was prepared as described above . All data was normalized to the housekeeping gene B2M5 [58] . Amplification and takeoff values were analyzed and graphed by REST2009 to determine any significant expression differences in DLX5 and DLX6 transcript levels between control and affected cDNA samples [24] . The case cohort consisted of 92 samples from individuals with nonsyndromic cleft lip with cleft palate ( NSCLP ) from the US , 59 samples from individuals with nonsyndromic cleft palate ( NSCP ) from the US and the Philippines , and 46 samples from individuals with cleft palate syndromes . The 46 syndromic samples consisted of 30 samples from individuals with PRS and 16 samples from individuals with additional congenital anomalies including club foot , hearing loss , heart disease , and intellectual disability . 60 unrelated CEPH samples ( CEU ) were used as controls . Primers were designed with Primer3 to cover the complete gene region of DLX6 and all exons of DLX5 [55] . Primer sequences and annealing temperatures are available in Table S5 . The first exon of DLX6 was sequenced using internal primers . PCR products were sequenced on an ABI 3730XL ( Functional Biosciences , Inc . , Madison , WI ) . Chromatograms were transferred to a Unix workstation , base-called with PHRED ( v . 0 . 961028 ) , assembled with PHRAP ( v . 0 . 960731 ) , scanned by POLYPHRED ( v . 0 . 970312 ) , and viewed with the CONSED program ( v . 4 . 0 ) . The functional effects of variants were predicted using the Variant Effect Predictor [59] . Genotypes from European or Asian populations available from the 1000 Genomes Project and the NHLBI Exome Sequencing Project were used as controls [25] , [26] . Fisher's exact test , implemented in STATA ( v12 . 1 ) , was used to test for differences in allele frequencies between NSCLP and NSCP cases and controls [60] . For insertions or deletions where allele frequencies were unavailable , we sequenced a set of 60 unrelated CEPH samples . Three SNPs ( rs2272280 , rs3801290 , and rs3213654 ) were genotyped in 362 case-parent trios ( 1090 individuals ) with nonsyndromic cleft lip with or without cleft palate ( NSCL/P ) from the US using TaqMan SNP Genotyping Assays ( Life Technologies , Grand Island , NY ) on the ABI Prism 7900HT , and were analyzed with SDS 2 . 3 software ( Applied Biosystems , Foster City , CA ) . The Family Based Association Test ( FBAT ) in STATA ( v12 . 1 ) was used to test for association these NSCL/P case-parent trios [60] . | Cleft palate is one of the most commonly occurring birth defects in children , and yet its cause is not completely understood . In order to better understand cleft palate we have turned to man's best friend , the domestic dog . Common breeding practices have made the dog a unique animal model to help understand the genetic basis of naturally occurring birth defects . A genome-wide association study of Nova Scotia Duck Tolling Retrievers with naturally occurring cleft palate led to the investigation of two homeobox genes , DLX5 and DLX6 . Dogs with this mutation also have a shortened lower jaw , which resembles those who have Pierre Robin Sequence ( PRS ) . Investigation into people with PRS identifies a mutation within a highly conserved and functional region of DLX5 that may contribute to the development of PRS . This exemplifies how the dog will help us better understand common birth defects . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genome-wide",
"association",
"studies",
"developmental",
"biology",
"animal",
"genetics",
"organism",
"development",
"genome",
"analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"molecular",
"genetics",
"computational",
"biology",
"genetics",
"of",
"disease",
"veterinary",
"science",
"veterinary",
"medicine"
] | 2014 | A LINE-1 Insertion in DLX6 Is Responsible for Cleft Palate and Mandibular Abnormalities in a Canine Model of Pierre Robin Sequence |
By aggregating data for complex traits in a biologically meaningful way , gene and gene-set analysis constitute a valuable addition to single-marker analysis . However , although various methods for gene and gene-set analysis currently exist , they generally suffer from a number of issues . Statistical power for most methods is strongly affected by linkage disequilibrium between markers , multi-marker associations are often hard to detect , and the reliance on permutation to compute p-values tends to make the analysis computationally very expensive . To address these issues we have developed MAGMA , a novel tool for gene and gene-set analysis . The gene analysis is based on a multiple regression model , to provide better statistical performance . The gene-set analysis is built as a separate layer around the gene analysis for additional flexibility . This gene-set analysis also uses a regression structure to allow generalization to analysis of continuous properties of genes and simultaneous analysis of multiple gene sets and other gene properties . Simulations and an analysis of Crohn’s Disease data are used to evaluate the performance of MAGMA and to compare it to a number of other gene and gene-set analysis tools . The results show that MAGMA has significantly more power than other tools for both the gene and the gene-set analysis , identifying more genes and gene sets associated with Crohn’s Disease while maintaining a correct type 1 error rate . Moreover , the MAGMA analysis of the Crohn’s Disease data was found to be considerably faster as well .
In the past decade , genome-wide association studies ( GWAS ) have successfully identified new genetic variants for a wide variety of phenotypes [1] . However , despite growing sample sizes , the genetic variants discovered by GWAS generally account for only a fraction of the total heritability of a phenotype [2 , 3] . More than anything , GWAS has shown that many phenotypes , such as height [4] , schizophrenia [5] and BMI [6] are highly polygenic and influenced by thousands of genetic variants with small individual effects , requiring very large sample sizes to detect them . Gene and gene-set analysis have been suggested as potentially more powerful alternatives to the typical single-SNP analyses performed in GWAS [7] . In gene analysis , genetic marker data is aggregated to the level of whole genes , testing the joint association of all markers in the gene with the phenotype . Similarly , in gene-set analysis individual genes are aggregated to groups of genes sharing certain biological , functional or other characteristics . Such aggregation has the advantage of considerably reducing the number of tests that need to be performed , and makes it possible to detect effects consisting of multiple weaker associations that would otherwise be missed . Moreover , gene-set analysis can provide insight into the involvement of specific biological pathways or cellular functions in the genetic etiology of a phenotype . Gene-set analysis methods can be subdivided into self-contained and competitive analysis , with the self-contained type testing whether the gene set contains any association at all , and the competitive type testing whether the association in the gene set is greater than in other genes [7] . Various methods for gene and gene-set analysis are currently available [7–13] . However , one concern with most existing methods is that they first summarize associations per marker before aggregating them to genes or gene sets . As demonstrated by Moskvina et al . this makes the statistical power strongly dependent on local linkage disequilibrium ( LD ) [14] , and also reduces power to detect associations dependent on multiple markers . An additional concern is that current gene-set analysis methods generally use a permutation-based approach . These are often very computationally demanding , and since no parametric model is used it is often not made explicit which null hypothesis is being evaluated and what assumptions are made . This makes it more difficult to determine the properties of the analysis such as how the significance of a gene set relates to the significance of its constituent genes or whether the analysis corrects for a polygenic architecture . This complicates the interpretation of results and hampers comparison between results of different gene-set analysis methods . To address such issues we have developed MAGMA ( Multi-marker Analysis of GenoMic Annotation ) , a fast and flexible tool for gene and gene-set analysis of GWAS genotype data . MAGMA’s gene analysis uses a multiple regression approach to properly incorporate LD between markers and to detect multi-marker effects . The gene-set analysis is built as a distinct layer around this gene analysis , providing the flexibility to independently change and expand both the gene and the gene-set analysis . Both self-contained and competitive gene-set analyses are implemented using a gene-level regression model . This regression approach offers a generalized framework which can also analyse continuous gene properties such as gene expression levels as well as conditional analyses of gene sets and other gene properties , and which can be extended to allow joint and interaction analysis of multiple gene sets and other gene properties as well . More traditional gene analysis models are also implemented , for comparison and to provide analysis of SNP summary statistics . To evaluate the performance of MAGMA we have applied it to the Wellcome Trust Case-Control Consortium ( WTCCC ) Crohn’s Disease ( CD ) GWAS data-set [15] , using the MSigDB Canonical Pathways [16] for the gene-set analysis . Simulation studies were performed to verify type 1 error rates for MAGMA . The CD data set was then analysed using MAGMA and with five commonly used other tools for gene and gene-set analyses , specifically VEGAS [17] , PLINK [8] , ALIGATOR [9] , INRICH [10] and MAGENTA [12] . The results show that MAGMA has greater statistical power than the other methods , while also being considerably faster .
The gene-set analysis is divided into two distinct and largely independent parts . In the first part a gene analysis is performed to quantify the degree of association each gene has with the phenotype . In addition the correlations between genes are estimated . These correlations reflect the LD between genes , and are needed in order to compensate for the dependencies between genes during the gene-set analysis . The gene p-values and gene correlation matrix are then used in the second part to perform the actual gene-set analysis . The advantage of decoupling these two parts of the analysis in this manner is that each can be changed independently from the other , simplifying the development of changes and extensions to either part of the model . Moreover , since the second part only uses the output from the first part and not the raw genotype data they do not need to be performed at the same time or place , making it much more straightforward to perform multiple gene-set analyses on the same data or to analyse multiple data sets across a large-scale collaboration . The gene analysis in MAGMA is based on a multiple linear principal components regression [18] model , using an F-test to compute the gene p-value . This model first projects the SNP matrix for a gene onto its principal components ( PC ) , pruning away PCs with very small eigenvalues , and then uses those PCs as predictors for the phenotype in the linear regression model . This improves power by removing redundant parameters , and guarantees that the model is identifiable in the presence of highly collinear SNPs . By default only 0 . 1% of the variance in the SNP data matrix is pruned away . With Xg* the matrix of PCs , Y the phenotype and W an optional matrix of covariates the model can thus be written as Y=α0g1→+Xg*αg+Wβg+εg , where the parameter vector αg represents the genetic effect , βg the effect of the optional covariates , α0g the intercept and εg the vector of residuals . The F-test uses the null-hypothesis H0: αg=0→ of no effect of gene g on the phenotype Y , conditional on all covariates . This choice of gene analysis model is motivated by a balance of statistical and practical concerns . This multiple regression model ensures that LD between SNPs is fully accounted for . It also offers the flexibility to accommodate additional covariates and interaction terms as needed without changing the model . At the same time , since the F-test has a known asymptotic sampling distribution the gene p-values take very little time to compute , making the gene analysis much faster than permutation-based alternatives . The linear regression model is also applied when Y is a binary phenotype . Although this violates some assumptions of the F-test , comparison of the F-test p-values with p-values based on permutation of the F-statistic shows that the F-test remains accurate ( see ‘Supplemental Methods—Implementation Details’ ) . MAGMA therefore uses the asymptotic F-test p-values by default , though it also offers an option to compute permutation-based p-values using an adaptive permutation procedure . In addition , comparison with logistic regression models shows that the results of the linear model are effectively equivalent to that of the more conventional logistic regression model , but without the computational cost . To perform the gene-set analysis , for each gene g the gene p-value pg computed with the gene analysis is converted to a Z-value zg = Φ−1 ( 1 – pg ) , where Φ−1 is the probit function . This yields a roughly normally distributed variable Z with elements zg that reflects the strength of the association each gene has with the phenotype , with higher values corresponding to stronger associations . Self-contained gene-set analysis tests whether the genes in a gene-set are jointly associated with the phenotype of interest . As such , using this variable Z a very simple intercept-only linear regression model can now be formulated for each gene set s of the form Zs=β01→+εs , where Zs is the subvector of Z corresponding to the genes in s . Evaluating β0 = 0 against the alternative β0 > 0 yields a self-contained test , since under the self-contained null hypothesis that none of the genes is associated with the phenotype zg has a standard normal distribution for every gene g . Competitive gene-set analysis tests whether the genes in a gene-set are more strongly associated with the phenotype of interest than other genes . To test this within the regression framework the model is first expanded to include all genes in the data . A binary indicator variable Ss with elements sg is then defined , with sg = 1 for each gene g in gene set s and 0 otherwise . Adding Ss as a predictor of Z yields the model Z=β0s1→+Ssβs+ε . The parameter βs in this model reflects the difference in association between genes in the gene set and genes outside the gene set , and consequently testing the null hypothesis βs = 0 against the one-sided alternative βs > 0 provides a competitive test . Note that this is equivalent to a one-sided two-sample t-test comparing the mean association of gene-set genes with the mean association of genes not in the gene-set . Similarly , the self-contained analysis is equivalent to a one-sided single-sample t-test comparing the mean association of gene-set genes to 0 . It should be clear that in this framework , the gene-set analysis models are a specific instance of a more general gene-level regression model of the form Z=β01→+C1β1+C2β2+…+ε . The variables C1 , C2 , … , in this generalized gene-set analysis model can reflect any gene property , from the binary indicators used for the competitive gene-set analysis to continuous variables such as gene size and expression levels . Any transformations of , and interactions between , such gene properties can also be added . This generalized gene-set analysis model thus allows for testing of conditional , joint and interaction effects of any combination of gene sets and other gene properties . In practice , the competitive gene-set analysis implemented in MAGMA in fact uses such a generalized model by default , performing a conditional test of βs corrected for the potentially confounding effects of gene size , gene density and ( if applicable , e . g . in meta-analysis ) difference in underlying sample size , if such effects are present . This is achieved by adding these variables , as well as the log of these variables , as covariates to the gene-level regression model . The gene density is defined as the ratio of effective gene size to the total number of SNPs in the gene , with the effective gene size in turn defined as the number of principal components that remain after pruning . One complication that arises in this gene-level regression framework is that the standard linear regression model assumes that the error terms have independent normal distributions , i . e . ε~MVN ( 0→ , σ2I ) . However , due to LD , neighbouring genes will generally be correlated , violating this assumption . This issue can be addressed by using Generalized Least Squares approach instead , and assuming that ε~MVN ( 0→ , σ2R ) . In MAGMA , the required gene-gene correlation matrix R is approximated by using the correlations between the model sum of squares ( SSM ) of each pair of genes from the gene analysis multiple regression model , under their joint null hypothesis of no association . These correlations are a function of the correlations between the SNPs in each pair of genes and thus provide a good reflection of the LD , and since they have a convenient closed-form solution they are easy to compute ( see also ‘Supplemental Methods—Implementation Details’ ) . Note that for the self-contained analysis , the submatrix Rs corresponding to only the genes in the gene set is used instead of R . In addition , since the self-contained null hypothesis guarantees that all zg have a standard normal distribution , the error variance σ2 can be set to 1 . Since raw genotype data may not always be available for analysis , MAGMA also provides more traditional SNP-wise gene analysis models of the type implemented in PLINK and VEGAS . These SNP-wise models first analyse the individual SNPs in a gene and combine the resulting SNP p-values into a gene test-statistic , and can thus be used even when only the SNP p-values are available . Although evaluation of the gene test-statistic does require an estimate of the LD between SNPs in the gene , estimates based on reference data with similar ancestry as the data the SNP p-values were computed from has been shown to yield accurate results [17 , 19] . Two types of gene test statistics have been implemented in MAGMA: the mean of the χ2 statistic for the SNPs in a gene , and the top χ2 statistic among the SNPs in a gene . For the mean χ2 statistic , a gene p-value is then obtained by using a known approximation of the sampling distribution [20 , 21] . For the top χ2 statistic such an approximation is not available , and therefore an adaptive permutation procedure is used to obtain an empirical gene p-value . A random phenotype is first generated for the reference data , drawing from the standard normal distribution . This is then permuted , and for each permutation the top χ2 statistic is computed for every gene . The empirical p-value for a gene is then computed as the proportion of permuted top χ2 statistics for that gene that are higher than its observed top χ2 statistic . The required number of permutations is determined adaptively for each gene during the analysis , to increase computational efficiency . Further details can be found in ‘Supplemental Methods—SNP-wise gene analysis’ . The MAGMA SNP-wise models can also be used to analyse raw genotype data , in which case the raw genotype data takes the place of the reference data and the SNP p-values are computed internally . Gene-set analysis based on these SNP-wise models proceeds in the same way as the gene-set analysis based on the multiple regression gene analysis model . The gene p-values resulting from the analysis are converted to Z-values in the same way to serve as input for the gene-set analysis . Similarly , the gene-gene correlation matrix R is obtained using the same formula as with the multiple regression model , but using the reference data to compute it . A number of additional features has been implemented in MAGMA , more fully described in ‘Supplemental Methods—Extensions’ . Gene analysis can be expanded with a gene-environment interaction component , which can subsequently be carried over to the gene-set analysis . Options for aggregation of rare variants and for fixed-effects meta-analysis for both gene and gene-set analysis are also available . Efficient SNP to gene annotation and a batch mode for parallel processing are provided to simplify the overall analysis process . MAGMA is distributed as a standalone application using a command-line interface . The C++ source code is also made available , under an open source license . MAGMA can be downloaded from http://ctglab . nl/software/magma . To evaluate the performance of MAGMA , the WTCCC Crohn’s Disease ( CD ) GWAS data [15] in conjunction with both WTCCC control samples was used . The data was cleaned according to the protocol described by Anderson [22] , resulting in a sample of 1 , 694 cases and 2 , 917 controls with data for 403 , 227 SNPs . The European samples from the 1 , 000 Genomes data [23] and the HapMap 3 data [24] were used as reference data sets for the summary statistics gene analysis . SNPs were annotated to genes based on dbSNP version 135 SNP locations and NCBI 37 . 3 gene definitions . For the main analyses only SNPs located between a gene’s transcription start and stop sites were annotated to that gene , yielding 13 , 172 protein-coding genes containing at least one SNP in the CD data . An additional annotation using a 10 kilobase window around each gene was made , yielding 16 , 970 genes , to determine the effect of using a window on relative performance . These two gene annotations were used for all analyses , to ensure that differences in default annotation settings did not cloud the comparison between tools . The 1 , 320 Canonical Pathways from the MSigDB database [16] were used for the gene-set analysis . The relatively large number of gene sets and the fact that the MSigDB Canonical Pathways are drawn from a number of different gene-set databases ensures a wide variety of gene sets , which should prevent the results from being too dependent on the choice of gene-set database . The MAGMA gene analysis was performed on the raw CD data using the PC regression model ( MAGMA-main ) . Gene analyses with VEGAS and PLINK were performed using the mean SNP statistic for VEGAS and both the mean SNP statistic ( PLINK-avg ) and the top SNP statistic ( PLINK-top ) for PLINK . Pruning in PLINK was turned off for these analyses . An additional PLINK analysis using the mean SNP statistic with pruning set to its default ( PLINK-prune ) was performed as well . To facilitate the comparison , several additional SNP-wise gene-set analyses were performed in MAGMA with test-statistics matching those of PLINK-avg , PLINK-top and VEGAS: mean χ2 ( MAGMA-mean ) and top χ2 ( MAGMA-top ) on the raw CD data to match the two PLINK analyses , and mean χ2 using CD SNP p-values and with either HapMap reference data ( MAGMA-pval ) to match VEGAS or with 1 , 000 Genomes reference data ( MAGMA-pval-1K ) . The SNP summary statistics used for VEGAS and MAGMA-pval were computed using PLINK ‘--assoc’ . Gene-set analysis for MAGMA was performed based on the PC regression gene analysis model ( MAGMA-main ) as well as on the SNP-wise model with 1 , 000 Genomes reference data ( MAGMA-pval-1K ) . Several other analyses were performed for comparison: PLINK self-contained gene-set analysis without pruning ( PLINK-avg ) and with pruning ( PLINK-prune ) , as well as ALIGATOR , INRICH and MAGENTA competitive gene-set analysis . PLINK operates on raw genotype data , whereas all three competitive methods require only SNP p-values as input . No correction for stratification was used in any of the analyses except when explicitly specified . An overview of all analyses is given in Table 1 .
Simulation was used to assess the type 1 error rates , using permutations of the CD phenotype to obtain a global null distribution of no associated SNPs ( see ‘Supplemental Methods—Simulation Studies’ for details ) . For the gene analysis , type 1 error rates were found to be controlled at the nominal level of 0 . 050 for the PC regression model , the summary statistics analysis model , as well as the SNP-wise models ( Table S1 in S2 File ) . The type 1 error rates for the gene-set analysis were also found to be well controlled for both the self-contained and competitive test ( Table S2 in S2 File ) . For the competitive test an additional simulation using a polygenic null model was performed , with effects explaining a combined 50% of the phenotypic variance assigned to randomly selected SNPs . This polygenic type 1 error rate was also well controlled . The type 1 error rates for the self-contained analysis under the polygenic null model are also shown . These are considerably inflated because self-contained gene-set analysis by its definition is not designed to correct for polygenicity , illustrating the risk of performing self-contained analysis on polygenic phenotypes . The results of the gene analyses of the CD data are summarized in Table 2 , which shows the number of significant genes at a number of different p-value thresholds . Since the Type 1 error rates have been shown to be properly controlled these results can serve as a good indicator of the relative power of the different methods , and compared to simulation-based power estimates this has the advantage that no assumptions about the genetic causal model . From Table 2 it is clear that whereas the power of all the other methods is very similar , the MAGMA-main model shows a clear advantage over the rest . After Bonferroni correction , MAGMA-main found a total of 10 genome-wide significant genes , including the well-known CD genes NOD2 , ATG16L1 and IL23R [25 , 26] . This also indicates that although MAGMA can perform analysis of summary statistics , raw data analysis should always be preferred if possible . Specific implementation issues can be ruled out as the cause of the power difference since the PLINK and VEGAS analyses yield results highly similar to their matched MAGMA models ( S9 Fig ) , and using the pruning option in PLINK also has little effect on the overall results . This means that the difference must be due to the difference in the methods and test-statistics themselves . Comparing the MAGMA implementations of these models in Fig 1 , the mean χ2 and top χ2 approaches are shown to produce very similar p-values . Moreover , the plots reveal that the superior power of the MAGMA-main model does not arise from consistently lower gene p-values , but rather from a small set of genes with low p-values for MAGMA-main that are simply not picked up by the other approaches . This is likely to be related to the way LD between SNPs is handled , as that is one of the key differences between the multiple regression model of MAGMA-main and all the others . A post-hoc power simulation indeed indicates that multi-marker effects with weak marginals are the most probable explanation ( see ‘Supplemental Methods—Simulation Studies’ ) . To increase the generalizability of these findings , two variations on the CD analyses were performed for MAGMA-main , MAGMA-mean and MAGMA-top . First , the analyses were repeated with 10 principal components computed from the whole data set as covariates to correct for possible stratification . The results are shown in Table 2 and S10 Fig . There is shown to be only very limited stratification , and although the power does decrease somewhat MAGMA-main’s power advantage is maintained . The analyses were also repeated with the gene annotation extended to include a 10 kilobase window around each gene , with the comparison in S11 Fig showing a considerable impact on the results . However , although this suggests that the choice of window can strongly affect the results of a gene analysis Table 2 shows that the relative power stays the same , with MAGMA-main again maintaining its superior power . As with the gene analysis , the results of the CD analysis ( Table 3 and Fig 2 ) can again serve as a gauge of the relative power of the different gene-set analysis methods . For the self-contained gene-set analysis this comparison is straightforward with MAGMA showing considerably more power than the two PLINK analyses . For the most part MAGMA’s power advantage can be explained by the difference in the underlying gene model , given the superior power of the PC regression model over the SNP-wise model used by PLINK shown before . Differences in how the genes are combined may also play a role however since , in contrast to PLINK , MAGMA weighs genes equally rather than by the number of SNPs in them and explicitly takes correlations between genes into account . Of note is also that PLINK-prune does considerably better than PLINK-avg , and that its p-values are somewhat more strongly correlated with those of the MAGMA analysis ( Fig 2 ) . An additional summary statistics analysis ( MAGMA-pval-1K ) on SNP p-values and using 1 , 000 Genomes reference data was also performed . This showed less power than PLINK even though it uses the same model at the gene level , suggesting that the difference is due to how the genes are aggregated to gene-sets . One of the key differences in this regard is that PLINK gives larger genes greater weight whereas MAGMA weighs them equally . As such a likely explanation is that the PLINK results are partially driven by a smaller number of large genes , though constructing the intermediate models to verify this is beyond the scope of this paper . The comparison of competitive methods is somewhat more complicated , due to the fact that ALIGATOR , INRICH and MAGENTA all use discretization using a p-value cut-off . This cut-off needs to be specified by the user and has no obvious default value , although for MAGENTA the 5th percentile cut-off is suggested as the most optimal [12] . For ALIGATOR and INRICH the analysis was therefore performed at four different cut-offs ( 0 . 0001 , 0 . 001 , 0 . 005 , 0 . 01 ) , and for MAGENTA at two ( 5th and 1st percentile ) . Of the four tools , only MAGMA and INRICH yield significant results after multiple testing correction ( Tables 3 and 4 ) . As with the self-contained gene-set analysis , power for the MAGMA analysis is better when using raw data rather than SNP p-values as input , though both yield one significant gene set . For INRICH the results are strongly dependent on the SNP p-value cut-off used , with three significant gene sets at the 0 . 0001 cut-off but none at the higher ones , further emphasizing the problem of choosing the correct cut-off . It should also be noted that the p-values have not been corrected for the fact that the gene-sets have been analysed under four different thresholds , and thus might not fall below the significance threshold if they were . Looking at the results in more detail ( Fig 3 ) also suggests that the differences in results are not merely due to a difference in power . The concordance between methods is poor , with only MAGENTA and ALIGATOR showing a reasonable correlation in results . Moreover , there is considerable discordance between different p-values cut-offs for the same methods as well ( Fig 4 ) . This suggests that the different methods , or methods at different p-value cut-offs , are sensitive to distinctly different kinds of gene set associations . In particular , MAGMA and the other three methods at higher p-value cut-offs would be expected to respond best to gene-sets containing a larger number of somewhat associated genes . Conversely , at lower p-value cut-offs the latter three should become more sensitive to gene-sets containing a small number of more strongly associated genes . This is exemplified by the INRICH analysis . At the 0 . 0001 cut-off only quite strongly associated genes are counted as relevant , but as there are only 42 such genes overall the three gene sets ( containing either 26 or 29 genes ) become significant despite each containing only three relevant genes . Aside from differences between methods , Table 3 also shows a clear difference between self-contained and competitive gene-set analysis . This is not a difference in power , but rather a difference of null hypothesis . Competitive tests attempt to correct for the baseline level of association present in the data and accordingly have a much more general null hypothesis . The impact of this difference in hypothesis can be illustrated by comparing the MAGMA self-contained and competitive analyses , since they are performed in the same framework . Whereas the self-contained analysis detects 39 gene sets that show association with the phenotype , the competitive analysis detects only one of those 39 . For the remaining 38 gene sets , there is no evidence in the data that the associations in those gene sets are any stronger than would be expected by chance given the polygenic nature of CD . The gene-set that remains is the Regulation of AMPK via LKB1 ( REACTOME ) set . For two additional gene sets , Cell Adhesion Molecules ( KEGG ) and ECM-receptor Interaction ( KEGG ) , the competitive p-value also drops below the significance threshold ( Table 4 and S12 Fig ) if the correction for gene size and gene density is turned off . This suggests that these gene sets do in fact contain significantly elevated levels of association , but that this is partially caused by confounding effects of the size and density of the genes they contain . Given the strength of the confounding effect it is evident that gene-set analyses should always be corrected for these and other potential confounders , to avoid false positive results . Full results for the analyses can be found in Table S5 in S2 File . All analyses were performed on the Genetic Cluster Computer , which is part of the Dutch Lisa Cluster . In terms of computational performance MAGMA is shown to have a considerable advantage over the other methods ( Table 5 ) for both gene and gene-set analysis . The most marked difference is between MAGMA and PLINK , the only one of the alternative methods using raw data input . However , the raw data analysis in MAGMA outperforms the summary statistics methods as well . Although INRICH and ALIGATOR show comparable computation times at their lowest SNP p-value cut-off , the need to repeat the analysis at multiple cut-offs means the total analysis for both takes considerably longer . The low MAGMA computation times are largely due to the choice of statistical model . Since the statistical tests used have known asymptotic sampling distributions the need for computationally demanding permutation or simulation schemes is avoided . Note however that the permutation-based SNP-wise analyses in MAGMA also show very reasonable computation times . These results demonstrate that , given efficient implementation , there is no computational reason to prefer analysis of summary statistics over raw data analysis , even when using permutation .
We have developed MAGMA , a fast and flexible method for performing gene and gene-set analysis in a two-tiered parametric framework . Comparison with a number of other , frequently used methods shows that MAGMA has better power for gene analysis as well as for both self-contained and competitive gene-set analysis . An important factor in this is the multiple regression model used in the gene analysis , which is better able to incorporate the LD between SNPs than other methods . Because of its two-layer structure , this improvement in power at the gene-level subsequently carries over to the gene-set analysis . MAGMA was also found to be generally much faster than other methods , even methods that used only summary statistics rather than raw data . This is primarily due to the choice of statistical model , which did not require the kind of computationally expensive permutation or sampling procedures used in the other methods . However , even the permutation-based SNP-wise models implemented in MAGMA outperformed their equivalents in other software and yielded very reasonable computation times . Although MAGMA showed better power than other tools for both the self-contained and competitive gene-set analysis , these comparisons also revealed considerable differences between the methods . This was most pronounced for the competitive gene-set analysis , with even results for individual methods showing significant variability based on the choice of cut-off . At present no comprehensive evaluation of the differences between existing gene-set analysis methods exists , leaving the causes and implications of these difference unclear . It is beyond the scope of this paper to perform such an evaluation , but the degree of discordance between most methods strongly suggests a need for future research in this direction . An additional caveat is that it is unknown to what extent the observed differences in power between methods may depend on the specific genetic architecture of Crohn’s diseases , and as such generalizing the results to other genetic architectures must be done with caution . The framework for MAGMA is built with future extensions in mind . Because of the two-tiered structure of the gene-set analysis , alternative gene analysis models are straightforward to implement and are automatically available for use in the gene-set analysis . Similarly , the linear regression structure used to implement the gene-set analysis offers a high degree of extensibility . At present it enables analysis of continuous gene-level covariates as well as conditional analysis of gene-sets correcting for possible confounders , and the analysis of the CD data demonstrates that correction for confounders such as gene size and gene density is indeed strongly advised . The model is easily generalized to much more general gene-level linear regression models to allow for simultaneous analysis of multiple covariates and gene-sets , opening up a wide range of new testable hypotheses . | Gene and gene-set analysis are statistical methods for analysing multiple genetic markers simultaneously to determine their joint effect . These methods can be used when the effects of individual markers is too weak to detect , which is a common problem when studying polygenic traits . Moreover , gene-set analysis can provide additional insight into functional and biological mechanisms underlying the genetic component of a trait . Although a number of methods for gene and gene-set analysis are available however , they generally suffer from various statistical issues and can be very time-consuming to run . We have therefore developed a new method called MAGMA to address these issues , and have compared it to a number of existing tools . Our results show that MAGMA detects more associated genes and gene-sets than other methods , and is also considerably faster . The way the method is set up also makes it highly flexible . This makes it suitable as a basis for more general statistical analyses aimed at investigating more complex research questions . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | MAGMA: Generalized Gene-Set Analysis of GWAS Data |
As miRNAs are associated with normal cellular processes , deregulation of miRNAs is thought to play a causative role in many complex diseases . Nevertheless , the precise contribution of miRNAs in fibrotic lung diseases , especially the idiopathic form ( IPF ) , remains poorly understood . Given the poor response rate of IPF patients to current therapy , new insights into the pathogenic mechanisms controlling lung fibroblasts activation , the key cell type driving the fibrogenic process , are essential to develop new therapeutic strategies for this devastating disease . To identify miRNAs with potential roles in lung fibrogenesis , we performed a genome-wide assessment of miRNA expression in lungs from two different mouse strains known for their distinct susceptibility to develop lung fibrosis after bleomycin exposure . This led to the identification of miR-199a-5p as the best miRNA candidate associated with bleomycin response . Importantly , miR-199a-5p pulmonary expression was also significantly increased in IPF patients ( 94 IPF versus 83 controls ) . In particular , levels of miR-199a-5p were selectively increased in myofibroblasts from injured mouse lungs and fibroblastic foci , a histologic feature associated with IPF . Therefore , miR-199a-5p profibrotic effects were further investigated in cultured lung fibroblasts: miR-199a-5p expression was induced upon TGFβ exposure , and ectopic expression of miR-199a-5p was sufficient to promote the pathogenic activation of pulmonary fibroblasts including proliferation , migration , invasion , and differentiation into myofibroblasts . In addition , we demonstrated that miR-199a-5p is a key effector of TGFβ signaling in lung fibroblasts by regulating CAV1 , a critical mediator of pulmonary fibrosis . Remarkably , aberrant expression of miR-199a-5p was also found in unilateral ureteral obstruction mouse model of kidney fibrosis , as well as in both bile duct ligation and CCl4-induced mouse models of liver fibrosis , suggesting that dysregulation of miR-199a-5p represents a general mechanism contributing to the fibrotic process . MiR-199a-5p thus behaves as a major regulator of tissue fibrosis with therapeutic potency to treat fibroproliferative diseases .
Tissue fibrosis , defined as the excessive and persistent formation of non functional scar tissue in response to repeated injury and insult , is a leading cause of morbidity and mortality associated with organ failure in various chronic diseases such as those affecting the lung interstitium [1] . Among the interstitial lung diseases of unknown etiology , Idiopathic Pulmonary Fibrosis ( IPF ) is the most common and lethal with a median survival of 3 to 5 years after diagnosis [2] . The pathogenesis of IPF is complex and largely unknown [2] , but observations based on both animal models of pulmonary fibrosis and lung sections from patients with IPF suggest a dynamic pathobiological process involving excessive wound healing with chronic inflammation , apoptosis of epithelial and endothelial cells , mesenchymal cell proliferation and activation with the formation of fibroblasts/myofibroblasts foci , and finally excessive deposition of extracellular matrix resulting in the destruction of the lung architecture and the loss of lung functions [2] . In particular , myofibroblasts play a substantial role in IPF by secreting important amount of ECM components and by promoting lung tissue stiffening [3] . Given the poor response rate of IPF patients to current therapy , a detailed understanding of the underlying pathogenic mechanisms is of major interest to develop new effective therapeutic strategies targeting the cellular and molecular events involved in the fibrotic response . MicroRNAs ( miRNAs ) are a class of noncoding small RNA , which most often bind to the 3′ UTR of target genes mRNAs and thereby repress their translation and/or induce their degradation . Since the first miRNA identification in Caenorhabditis elegans in a context of larval development [4] , [5] , thousands miRNAs have now been characterized including about 2000 in human ( miRbase v19 ) [6] . MiRNAs are now recognized as major regulators of gene expression with crucial functions in numerous biological processes including development , proliferation , differentiation , apoptosis and stress response . Importantly , recent studies have identified specific miRNA expression patterns related to the initiation and progression of various diseases including cancer as well as inflammatory , infectious and autoimmune diseases [7]–[9] . Additionally , gain and loss of function miRNA studies have further established their functional impact in various in vivo models [10]–[15] . Nevertheless , the precise contribution of miRNAs in fibrotic diseases , especially lung fibrosis , is still poorly understood [16] , [17] . Our rationale was therefore to test whether miRNAs may provide new perspectives on disease mechanisms , diagnosis as well as new therapeutic opportunities in the specific context of fibrosis . In an effort to identify miRNAs with potential roles in the development of lung fibrosis ( strategy detailed in Figure S1 ) , we aimed to identify miRNAs of interest in two mouse strains showing different susceptibility to develop lung fibrosis after bleomycin exposure . This led to the identification of a panel of miRNAs specifically dysregulated in the lungs of fibrosis prone mouse strain in response to bleomycin . Among these miRNAs , miR-199a-5p was found to be selectively up-regulated in myofibroblasts of the injured lung in bleomycin-treated mice and fibroblastic foci of IPF patients . In lung fibroblasts , miR-199a-5p acts as an effector of TGFβ signaling , regulates CAV1 expression , a critical mediator of the lung fibrosis process [18]–[21] and participates to multiple fibrogenic associated-processes including cell proliferation , migration , invasion and differentiation into myofibroblasts . Finally , dysregulation of miR-199a-5p was also found in two other mouse models of tissue fibrosis , namely kidney fibrosis and liver fibrosis , suggesting therefore that miR-199a-5p is likely to be a common mediator of fibrosis .
Previous studies based on mice have demonstrated a genetic susceptibility to bleomycin-induced pulmonary fibrosis [22] , [23] . Indeed , C57BL/6 mice are considered to be fibrosis prone , whereas BALB/C mice are less prone to fibrosis . To identify miRNAs that may contribute to the lung fibrosis process , miRNA expression profile in response to bleomycin was assessed 7 days and 14 days following bleomycin administration ( i . e . when active fibrogenesis occurs ) on both strains using a microarray based platform ( Data set 1 , GEO accession number GSE34812 ) described elsewhere [24]–[26] . We identified 22 differentially expressed miRNAs between lungs from bleomycin- and control-treated animals in at least one strain , the majority being upregulated in bleomycin-instillated lungs ( Figure 1A ) . We focused our analysis on miRNAs that exhibited an enhanced expression in response to bleomycin during disease progression in the C57BL/6 sensitive mice only . Among several miRNAs candidates with such a profile , miR-199a-5p displayed the highest statistical score ( Figure 1B ) . This was further established using an independent set of mice at day 14 following bleomycin treatment ( Figure S2A ) . These findings strongly suggested that miR-199a-5p may play an important role during the lung fibrosis process . To investigate the regulatory mechanisms underlying miR-199a-5p production , we assessed the expression status of the 2 mouse genes , miR-199a-1 ( on chromosome 9 ) and miR-199a-2 ( on chromosome 1 ) in response to bleomycin using a Taqman assay designed to discriminate between pri-miR-199a-1 and pri-miR-199a-2 . Our results showed that , 14 days after bleomycin instillation , both pri-miR-199a transcripts were up-regulated in the lungs of C57BL/6 mice ( Figure S2B ) and thus , contributed to miR-199a-5p production . In addition , in situ hybridization experiments performed in the injured lungs from C57BL/6 mice 14 days after bleomycin instillation revealed a selective expression of miR-199a-5p in myofibroblasts ( Figure 1C ) . Of note , consistent with previous findings [14] , we also found a significant upregulation of miR-21 ( now referenced in miRbase as mmu-miR-21a-5p ) in response to bleomycin ( Figure 1A and Figure S3 ) . Nevertheless , miR-21 induction did not differ between bleomycin-sensitive and bleomycin-resistant strains of mice . We next sought to determine the mechanism by which miR-199a-5p dysregulation may lead to tissue fibrosis . To address this question , we first attempted to identify gene targets and cellular pathways regulated by miR-199a-5p using the methodology described earlier [25] , [26] . The influence of miR-199a-5p on human pulmonary hFL1 fibroblast transcriptome was compared with that of miR-21 , which has been previously associated with the development of fibrotic diseases including lung fibrosis [14] , [15] , [27] ( Data set 2 , GEO accession number GSE34815 ) . Forty-eight hours after ectopic overexpression of each miRNA , a significant alteration ( defined by an absolute log2ratio above 0 . 7 and an adjusted p-value below 0 . 05 ) of 1261 and 753 transcripts was detected in the miR-199a-5p and miR-21 conditions , respectively . While these 2 miRNAs induced very distinct gene expression patterns ( Figure 2A ) , a functional annotation of these signatures , using Ingenuity Pathway software , indicated an overlap for “canonical pathways” including “Cell Cycle regulation” and “TGFβ Signaling” ( Table S1 ) . Consistent with previous findings [28] , highly significant pathways associated with miR-21 were related to “Cyclins and Cell Cycle Regulation” as well as “Cell Cycle Control of Chromosomal Replication” , “Mismatch Repair in Eukaryotes” and “ATM signalling” . While the highest scoring pathway for miR-199a-5p corresponded to the metabolic pathways “Biosynthesis of Steroids” , we also noticed enrichment for pathways related to “Integrin Signaling” and “Caveolar-mediated Endocytosis Signaling” . We next looked for an enrichment of putative direct targets in the population of down-regulated transcripts , as described in [29] . A specific overrepresentation of predicted targets for miR-199a-5p and miR-21 in the population of down-regulated transcripts was noticed after heterologous expression of either miR-199a-5p or miR-21 , respectively . This enrichment was independent of the prediction tool used to define the targets ( Figure 2B and not shown ) . We then focussed our analysis on a subset of 21 transcripts containing miR-199a-5p complementary hexamers in their 3′UTR , showing the largest inhibition of expression , and identified by TargetScan , PicTar and miRanda ( Figure 2C and Table 1 ) . The gene list of interest was further narrowed by focussing on targets also associated with the most significant canonical pathways described above . Interestingly , the expression levels of 4 out of 21 mouse orthologs were also significantly down-regulated in C57BL/6 mice 14 days after instillation of bleomycin ( Data set 3 , GEO accession number GSE34814 , Table S2 ) . These targets , highlighted in Table 1 , are ARHGAP12 , CAV1 , MAP3K11 and MPP5 . Based on previous studies that demonstrated a significant link between the downregulation of caveolin-1 ( CAV1 ) in lung fibroblasts and the deleterious effects mediated by TGFβ [19] , [30] , CAV1 represented a particularly relevant putative miR-199a-5p target gene . Alignment of miR-199a-5p with human CAV1 3′UTR sequence revealed one potential conserved seed site ( Figure 3A ) . We then fused part of the human CAV1 3′UTR to a luciferase reporter using the psiCHEK-2 vector and transfected it into HEK293 cells in the presence of either a pre-miR-199a-5p mimic or a pre-miR-control ( Figure 3B ) . As a control , we also used a CAV1 3′UTR construct mutated on the predicted miR-199a-5p site . Human pre-miR-199a-5p induced a significant decrease in the normalized luciferase activity relative to control in the presence of the wild type construction only , confirming that it represents a functional site . Moreover , this inhibition was also repeated using the whole 3′-UTR of human CAV1 ( Figure S4 ) , demonstrating that CAV1 is indeed a direct target of miR-199a-5p . Finally , transfection of pre-miR-199a-5p into MRC-5 and hFL1 lung fibroblasts led to a significant and specific decrease of CAV1 at both mRNA and protein levels while miR-21 had no significant effect ( Figure 3C–3E and Figure S5 ) . As TGFβ is known to downregulate CAV1 in pulmonary fibroblasts [19] , we then investigated whether decreased expression of CAV1 upon TGFβ stimulation was associated with an increase in miR-199a-5p expression . We exposed the MRC-5 cell line to TGFβ , and analyzed the expression levels of CAV1 and miR-199a-5p . As detected by Taqman RT-PCR , TGFβ treatment of human fibroblasts for 24 h or 48 h caused a marked decrease of CAV1 mRNA , whereas miR-199a-5p expression was significantly upregulated ( Figure 4A and 4B ) . Decrease of CAV1 protein levels after TGFβ treatment was time dependent ( Figure 4C ) . To further investigate whether miR-199a-5p is involved in TGFβ-induced downregulation of CAV1 , we performed additional experiments using a LNA-based inhibitor of miR-199a-5p as well as a CAV1 target site blocker to specifically interfere with miR-199a-5p binding on CAV1 3′UTR . As depicted in Figure 4D and S6 , both LNA-mediated silencing of miR-199a-5p and blocking miR-199a-5p binding on CAV1 3′UTR inhibit TGFβ-induced downregulation of CAV1 . Altogether , these experiments demonstrate that , in lung fibroblasts , induction of miR-199a-5p in response to TGFβ mediates CAV1 downregulation through binding on a unique site located in CAV1 3′UTR . We then assessed the expression of CAV1 in the fibrotic lungs of mice . Consistent with previous studies [19] , [31] , our data showed a significant decrease in both CAV1 mRNA and protein expression levels in C57Bl/6 mice 14 days after bleomycin administration ( Figure 5A–5C ) . Additionally , immunohistochemistry staining of CAV1 on lung tissue sections from C57Bl/6 mice 14 days after bleomycin treatment revealed a marked reduction of CAV1 in fibrotic area of the lungs ( Figure 5D ) . Taken together , these experiments show that the observed up-regulation of miR-199a-5p expression in the fibrotic lungs of mice is correlated with a downregulation of CAV1 . Of note , BALB/c mice , for which pulmonary expression of miR-199a-5p was not upregulated in response to bleomycin , did not display a significant decrease in CAV1 mRNA expression level 14 days after bleomycin treatment ( Figure S7 ) . Expression of miR-199a-5p expression was increased in lungs of IPF patients ( GEO accession number GSE13316 from [13]; dataset consisting of ten IPF samples and ten control samples; two different probes for miR-199a-5p with a p-value of p = 0 . 005 and p = 0 . 006 , wilcoxon rank sum test , Table S3 ) . This result was confirmed with an independent dataset composed of 94 IPF and 83 control lungs ( p<0 . 001 ) ( Figure 6A ) and in an additional cohort using qPCR ( Figure S8 ) . As observed in mice , IPF samples also exhibited a significant decrease in CAV1 expression ( p<0 . 001 ) ( Figure 6B ) . The linear fold ratio for CAV1 between IPF and control was 0 . 54 ( FDR<0 . 05 ) and the linear fold ratio for miR-199a-5p for the same subjects was 1 . 35 ( p<0 . 05 ) . Finally , examination of IPF lung sections revealed a specific expression of miR-199a-5p in fibroblastic foci of the injured lung as well as a decreased CAV1 expression ( Figure 6C and 6D ) . Given that loss of CAV1 expression represents a critical factor involved in the fibrogenic activation of pulmonary fibroblasts [19] , we assessed whether overexpression of miR-199a-5p in lung fibroblasts was sufficient to recapitulate known profibrotic effects associated with a decrease in CAV1 expression ( i . e . ECM synthesis , fibroblasts proliferation , migration , invasion and differentiation into myofibroblasts ) [30] , [32] , [33] . Transfection of miR-199a-5p precursors resulted in a significant induction of migration ( Figure 7A and 7B ) and invasion ( Figure 7C ) . In addition , cell cycle analysis ( percent cells in S phase ) showed that proliferation rate of pulmonary fibroblasts overexpressing miR-199a-5p was significantly enhanced ( Figure 7D ) . Finally , heterologous expression of miR-199a-5p also led to a strong increase in α smooth mucle actin ( αSMA ) expression ( Figure 7E and Figure S9 ) , a hallmark of myofibroblast differentiation as well as to a significant potentiation of COL1A1 induction in response to TGFβ ( Figure 7F ) . Comparison of the gene expression profiles obtained in lung fibroblasts transfected with miR-199a-5p precursors or with a siRNA specifically directed against CAV1 revealed an overlap between the 2 signatures , mainly among the down-regulated transcripts ( Figure S10A , group 2 ) : 34% of miR-199a-5p downregulated transcripts were also repressed by a siCAV1 ( Figure S10B ) . To gain insights into the pathways modulated by miR-199a-5p , Ingenuity Pathways canonical pathways associated to miR-199a-5p were analyzed and compared to those of miR-21 and siCAV1 conditions . This analysis revealed some proximity between miR-199a-5p and siCAV1 based on the existence of shared regulated pathways ( Figure 8A ) . Pathways that were specific to miR-199a-5p were related to inflammation , such as “IL-1 Signaling” , “Acute Phase Response Signaling” and “P38 MAPK Signaling” , i . e . all typical of fibrotic processes . Importantly , several profibrotic genes were specifically regulated by miR-199a-5p and their altered expression was confirmed in vivo ( Figure S11 and Table S4 ) . MiR-199a-5p thus regulates multiple signaling pathways involved in lung fibrogenesis . In particular , compared to siCAV1 transfected cells , overexpression of miR-199a-5p significantly increased CCL2 , TGFBRI and MMP3 expression and significantly decreased CAV2 and PLAU expression ( Figure S12 ) . Of note , these two downregulated genes were predicted to be direct targets of miR-199a-5p by Pictar . We next investigated whether miR-199a-5p is associated with TGFβ signaling . For this , we experimentally defined a TGFβ signaling signature in lung fibroblasts ( Dataset 2 , GSE34815 ) and compared it to miR-199a-5p signature using gene set enrichment analysis ( GSEA ) [34] . This analysis revealed a significant overlap between these two signatures , as assessed by normalized enrichment scores above 1 ( 1 . 4 and 2 . 17 for up- and down-regulated genes respectively , with nominal p-value and FDR q-value being <0 . 05 ) , suggesting therefore , that miR-199a-5p is involved in the TGFβ response of lung fibroblasts ( Figure 8B ) . To further demonstrate the importance of miR-199a-5p in TGFβ response , silencing of miR-199a-5p was performed in lung fibroblasts using LNA-based inhibitors . In particular , we showed that LNA-mediated silencing of miR-199a-5p strongly inhibited TGFβ-induced differentiation of lung fibroblasts into myofibroblasts ( Figure 8C and Figure S6 ) , SMAD signaling ( Figure 8D ) and stimulation of wound repair ( Figure 8E and 8F ) . Remarkably , similar results were obtained using CAV1 protector , demonstrating therefore that miR-199a-5p is a key effector of TGFβ response through CAV1 regulation ( Figure 8C , 8D , 8E , 8F and Figure S6 ) . A growing body of evidence suggests that miRNAs contribute to the fibrotic process in various organs such as heart , kidneys , liver or lungs . For example , previous reports have shown that miR-21 has an important role in both pulmonary and heart fibrosis experimental mouse models . Thus , we investigated whether miR-199a-5p was also dysregulated in other fibrotic tissues , namely kidney fibrosis and liver fibrosis . To this end , we assessed the overlap between the miRNA expression profiles corresponding to three experimental models of fibrosis . Measurements were made using the same miRNA based platform . We identified 5 miRNAs commonly dysregulated at a p-value of less than 0 . 01 ( Figure 9A ) . Among these miRNAs , 3 were downregulated ( miR-193 , miR-30b and miR-29c ) and 2 were upregulated ( miR-199a-3p and miR-199a-5p ) ( Figure 9B ) . The enhanced expression of miR-199a-5p was confirmed in two independent experimental models of liver fibrosis ( Figure 10A–10C ) and was correlated with the severity of liver fibrosis , as BALB/C mice have a more pronounced liver fibrosis than C57BL/6 mice , following administration of CCL4 ( Figure 10A and 10B ) . In addition , miR-199a-5p was significantly decreased during regression of experimental CCL4-induced liver fibrosis ( Figure 10D ) . Furthermore , we showed that TGFβ exposure of stellate cells was associated with an increase of miR-199a-5p expression and a decrease of CAV1 expression level ( Figure 10E and 10F ) . Interestingly , enhanced expression of miR-199a-5p was also observed in clinical samples from patients with liver fibrosis ( Figure S13 ) . Similarly , data obtained from the unilateral ureteral obstruction model of kidney fibrosis showed an enhanced expression of miR-199a-5p in the injured kidney compared to sham operated mice ( Figure 11A ) . Interestingly , as for lung fibrosis , kidney expression of miR-199a-5p was correlated with disease progression . As depicted in Figure 11B , in situ hybridization performed 28 days after surgery ( i . e . when the fibrosis is established ) showed no detectable signal for miR-199a-5p in normal kidney , whereas the hybridization signal was greatly enhanced throughout the injured kidney in area consistent with ( myo ) fibroblasts . Furthermore , immunohistochemistry of CAV1 performed on fibrotic kidney from mice 28 days after surgery showed a marked reduction of CAV1 in fibrotic area of the kidney ( Figure 11C ) .
MiRNA expression profiling using high-throughput genomic approaches has provided important new insights into the pathogenesis , classification , diagnosis , stratification , and prognosis of many human diseases including tissue fibrosis [15] , [35] , [36] . In particular , such approaches have been previously successfully applied to IPF , revealing miR-21 and let-7d as important contributors to the lung fibrosis process [13] , [14] . Our work represents however , to our knowledge , the first analysis of miRNAs involved into the differential susceptibility of two murine strains to bleomycin-induced lung fibrosis . The identification of a specific miRNA profile associated with bleomycin-sensitive animals suggests the functional importance of these dysregulated miRNAs during the pathogenic processes leading to lung fibrosis . MiR-199a-5p appeared as the most statistically significant and was well correlated to IPF progression . Thus , altered expression of miR-199a-5p is likely to represent a primary pathogenic mechanism in the development of lung fibrosis rather than a secondary effect of the long-standing disease process . Other miRNAs candidates including miR-214 , clustered with miR-199a-2 on mouse chromosome 1 as well as other miRNAs that have been previously associated to fibrosis , including miR-221-222 and miR-449a [37] , [38] also showed an enhanced expression in the lung fibrosis-susceptible mice . These miRNAs need to be further analyzed in IPF samples , as previous studies have shown their implication in the regulation of the stress response or the cell cycle/apoptosis balance in the epithelial or fibroblast compartment [38]–[41] . MiR-199a is an evolutionary conserved small RNA initially identified in the context of inner ear hair cells development and chondrogenesis [42]–[44] and numerous reports have now shown its implication in various tumor types [45]–[47] . In the context of tissue fibrosis , both mature forms of miR-199a ( i . e . , miR-199a-5p and miR-199a-3p ) have been associated with the progression of liver fibrosis in both humans and mice [48] , [49] . While our data also showed an enhanced pulmonary expression of these two miRNAs in the bleomycin-induced mouse model , expression of miR-199a-5p was more significant in IPF samples ( Table S3 and Figure S14A ) . Moreover , our data indicated that miR-199a-3p had distinct effects on lung fibroblasts differentiation than miR-199-5p , as assessed by their different impact on αSMA ( Figure S14B and data not shown ) . This led us to investigate in depth miR-199a-5p profibrotic effects . In a recent report describing the miRNA expression profile of lung fibroblasts , miR-199a-5p was found to be highly expressed [25] . Our present data establish stromal cells as the primary source of miR-199a-5p in the injured lungs and also suggests that miR-199a-5p is involved in the profibrotic effects mediated by pulmonary fibroblasts . A combination of in silico and experimental data , described in [25] , [26] , [40] , identified the transcripts affected by miR-199a-5p in lung fibroblasts . Functional annotations of the miR-199a-5p experiments highlighted terms such as “Integrin Signaling” and “Caveolar-mediated Endocytosis Signaling” . Among the set of transcripts that were down-regulated after ectopic expression of miR-199a-5p , we then restricted our work to a group of 21 miR-199a-5p target genes predicted by 3 independent algorithms , showing the largest modulation factors and smallest statistical p-values . This short list included CAV1 , a structural component of caveolae , previously associated with lung fibrosis [14] , [18] , [19] . Caveolae refer to 50–100 nanometers small bulb-shaped invaginations of the plasma membrane . They exert major biological functions in numerous cellular processes such as membrane trafficking or cell signaling [50] . CAV1 and CAV2 , the main coat proteins of caveolae , are relatively highly expressed in endothelial cells and fibroblasts of pulmonary origin [51] . Caveolae role is particularly important in the context of TGFβ signaling . Whereas TGFβ receptor endocytosis via clathrin-coated pit-dependent internalization promotes TGFβ signaling , the lipid raft-caveolar internalization pathway facilitates the degradation of TGFβ receptors , therefore decreasing TGFβ signaling [52] . Previous studies have shown that a reduced CAV1 expression in lung fibroblasts contributes to IPF pathogenesis by promoting TGFβ profibrotic effects [19] . In line with this , we provide evidence that miR-199a-5p can directly repress CAV1 in lung fibroblasts , thereby stimulating their proliferation , migration , invasion and differentiation into myofibroblasts ( Figure 12 ) . Additionally , we showed in a large cohort of IPF patients an enhanced expression of miR-199a-5p that was reproduced in three independent mouse models of fibrosis as well as a decreased expression of CAV1 . Finally , in contrast to a recent report showing that miR-199a-5p , by targeting SMAD4 , inhibited TGFβ-induced gastric cancer cell growth [53] , we found that lung fibroblasts overexpressing miR-199a-5p have an increased SMAD4 expression ( Figure S14B ) , suggesting thus a potential opposite function of this miRNA between epithelial and mesenchymal cells . MiRNAs , by affecting the expression of multiple genes , can act as master regulators of complex biological processes and aberrant expression of miRNA is known to have a profound impact on various distinct biological pathways . Thus , the elucidation of the critical genes and relevant pathways/networks modulated by miRNAs is important to understand the mechanisms by which miRNAs exert their pathogenic effects . Our systematic analysis of the gene expression profiles of lung fibroblasts overexpressing miR-199a-5p led to the identification of a large number of transcripts that were significantly modulated by this miRNA . These experiments have established that miR-199a-5p is directly or indirectly involved in the regulation of genes previously associated with lung fibrosis: CCL2 , a potent mononuclear cell chemoattractant , PLAU [54] , a component of the fibrinolysis system , TGFBRI , the TGFβ receptor type I [55] , MMP3 [56] and CAV2 [57] . Interestingly , these regulations were independent of CAV1 targeting , suggesting therefore that miR-199a-5p modulates the expression of several components of various distinct biological pathways to elicit , in lung fibroblasts , a profibrotic response . Before this study , miR-21 was clearly established as an effector of TGFβ signaling , able to promote fibroblast proliferation and differentiation into myofibroblasts [58] . In the context of lung fibrosis , miR-21 has been described to mediate lung fibroblast activation and fibrosis [14] . MiR-199a-5p and miR-21 exert indeed similar pro-fibrotic effects on lung fibroblasts . This is further demonstrated by overexpression of miR-21 and miR-199a-5p , which induce lung fibroblast migration to a similar extent ( Figure S15 ) . Interestingly , while both miRNAs appear as TGFβ effectors , the comparison of their associated gene expression signature indicated a limited overlap ( Figure 2A ) . Moreover , CAV1 expression is unaffected by overexpression of miR-21 in lung fibroblasts , suggesting that both miRNAs , in response to TGFβ , modulate distinct signaling pathways to produce cooperative effects involved in fibroblast activation . The mechanisms involved in the TGFβ-dependent modulation of miR-21 and miR-199a-5p are also of particular interest . While both miR-21 and miR-199a-5p have been shown to be regulated by TGFβ , their expression may be primarily regulated through a Smad-dependent post-transcriptional mechanism promoting miRNA maturation by Drosha [59] , [60] . Our data showing that both pri-miRNA-199a1 and pri-miRNA-199a2 are significantly upregulated in bleomycin-treated mice ( Figure S2B ) and TGFβ-stimulated fibroblasts ( Figure S16 ) suggest that additional TGFβ-dependent transcriptional regulations occur that will need to be more fully analyzed . Finally , our observation that miR-199a-5p is also dysregulated in two additional experimental models of tissue fibrosis ( i . e . kidney fibrosis and liver fibrosis ) establishes miR-199a-5p as a ubiquitous factor associated with tissue fibrogenesis . The recently reported association of CAV1 with kidney fibrosis [61] , [62] , together with the exclusive distributions of miR-199a-5p and CAV1 in the injured kidney , leads us to hypothesize that miR-199a-5p also controls CAV1 expression in kidney , thus contributing to kidney fibrosis . Further information came from the liver fibrosis model . As liver fibrosis can regress after cessation of the triggering injury , even at advanced fibrotic stages [63] , the decrease of miR-199a-5p observed during resolution of liver fibrosis sets for the first time a specific miRNA as an important player for orchestrating the molecular events occurring during regression of liver fibrosis . Importantly , this implies that therapeutic strategies based on modulation of miRNAs have a potential to prevent liver fibrosis progression but also to resolve liver fibrosis . In conclusion , the results of this study further underline the pivotal roles played by specific miRNAs in mediating changes in gene expression and cell functions occurring during pulmonary fibrosis . In particular , our results identified miR-199a-5p as a new determinant of tissue fibrosis . We thus anticipate that strategies preventing the up-regulation of miR-199a-5p may represent a new effective therapeutic option to treat fibroproliferative diseases .
Human normal pulmonary fibroblasts MRC-5 ( CCL-171 ) and hFL1 ( CCL-153 ) , human lung cancer cell line A549 ( CCL-185 ) and HEK-293 ( CRL-1573 ) cells were purchased from the American Type Culture Collection ( ATCC , Manassas , VA , USA ) , frozen at an early passage and each vial used for experiments was cultured for a limited number of passages ( <8 ) . For maintenance , cells were cultured in the appropriate medium ( MEM for MRC-5 , F12-K for hFL1 and A549 , DMEM for HEK-293 ) containing 10% fetal calf serum ( FCS ) , at 37°C with 5% v/v CO2 . Recombinant TGFβ was purchased from Sigma-Aldrich . The following monoclonal ( mAbs ) and polyclonal ( pAbs ) Antibodies were used: rabbit anti-CAV1 pAbs ( sc-894 , Santa Cruz Biotechnology Inc . ) , rabbit anti-SMAD4 ( 9515 ) and anti- β-Actin ( 13E5 ) mAbs ( Cell Signalling ) , mouse anti- αSMA mAbs ( 1A4 , Dako ) for immunohistochemistry and ( 4A8-2H3 , Abnova ) for Western Blot and immunocytofluorescence . All animal care and experimental protocols were conducted according to European , national and institutional regulations . Personnel from the laboratory carried out all experimental protocols under strict guidelines to insure careful and consistent handling of the mice . Primary stellate cells were isolated from C57BL/6 mice at the age of 40 to 55 weeks and stimulated with 20 ng/ml of TGF-ß ( Sigma Aldrich ) for 48 h as previously described [65] . Flash frozen lung tissue from 94 human subjects with IPF and 83 control subjects with no chronic lung disease were obtained from the Lung Tissue Research Consortium ( LTRC ) . These diagnoses were made using ATS/ERS guidelines [66] , [67] from review of clinical history , pathology , and radiology . All experiments were approved by the local Institutional Review Board at the University of Pittsburgh ( IRB# 0411036 ) . Clinical data were made entirely available to the investigators for review . Paraffin lung sections from patients with IPF were obtained from Lille's Hospital . Experiments were approved by the institutional review board of Lille's Hospital . Kidneys and lungs were fixed overnight with neutral buffered formalin and then embedded in paraffin . Five-micrometer-thick sections were mounted and stained with hematoxylin and eosin as well as Masson's trichrome to assess the degree of fibrosis . Histologic sections were reviewed by an experienced pathologist . Total RNA were extracted from lung tissue and cell samples with TRIzol solution ( Invitrogen ) . Integrity of RNA was assessed by using an Agilent BioAnalyser 2100 ( Agilent Technologies ) ( RIN above 7 ) . Cells or tissues were lysed in lysis buffer ( M-PER protein extraction reagent for cells , T-PER protein extraction reagent for tissues ) and protease inhibitors cocktail ( Pierce ) . The lysates were quantified for protein concentrations using the Bradford assay ( Biorad ) . Proteins ( 10 µg per sample ) were separated by SDS-polyacrylamide gel and transferred onto nitrocellulose membranes ( GE Healthcare ) . The membranes were blocked with 5% fat free milk in Tris-buffered saline ( TBS ) containing 0 . 1% Tween-20 ( TBS-T ) and subsequently incubated with CAV1 , α-SMA or β-actin primary antibodies overnight at 4°C . After washing with TBS-T for 30 minutes at room temperature , the membrane was further incubated with horseradish peroxidase–conjugated secondary antibodies for 1 . 5 hours , followed by 30 minutes of washing with TBS-T . Protein bands were visualized with Amersham ECL substrates ( GE Healthcare ) . Five-µm paraffin-embedded sections were sequentially incubated in xylene ( 5 minutes twice ) , 100% ethyl alcohol ( 5 minutes twice ) , 95% ethyl alcohol ( 5 minutes twice ) , and 80% ethyl alcohol ( 5 minutes ) . After washing with water , the sections were antigen-retrieved using citrate buffer ( pH 6 . 0; DAKO ) in a steamer for 20 minutes and cooled to ambient temperature . Sections were then washed with TBS-T and quenched with 3% hydrogen peroxide in TBS for 10 minutes , blocked for avidin/biotin activity , blocked with serum-free blocking reagent , and incubated with primary antibody as follows: for CAV1 staining , sections were incubated with antibody for 1 hour at ambient temperature; for alpha-SMA staining , sections were incubated with antibody overnight at 4°C . Immunohistochemical staining was developed using the DAB substrate system ( DAKO ) . In situ hybridization of miR-199a-5p was performed using double DIG-labeled LNA probes ( Exiqon , Woburn , MA ) . Paraffin-embedded mouse tissues were dewaxed in xylene and rehydrated in descending grades of alcohol . The slides were then washed in PBS ( pH 7 . 5 ) and permeabilized by incubating for 15 min in proteinase K ( Ambion ) for 15 min at 37°C . The slides were again washed in PBS , and prehybridized in hybridization buffer ( 50% formamide , 5× SSC , 0 . 1% Tween-20 , 9 . 2 mM citric acid , 50 µg/ml heparin , and 500 µg/ml yeast RNA , pH 6 ) in a humidified chamber . The double DIG-labeled LNA probes were then added to the sections at a 80 nM concentration and incubated 2 hours at 50°C in a humidified chamber . The slides were rinsed in 5× SSC , 1× SSC and 0 . 2× SSC solutions at the same hybridization temperature . This step was followed by blocking with 2% sheep serum , 2 mg/ml BSA in PBS+0 . 1% Tween-20 ( PBST ) and incubation with anti-DIG-AP Fab fragments antibody ( 1∶800 ) ( Roche Applied Sciences ) for 2 hours at room temperature . After washing in PBST , the color reaction was carried out by incubation in 5-bromo-4-chloro-3-indolyl phosphate ( BCIP ) /nitro blue tetrazolium ( NBT ) color solution ( Roche Applied Sciences ) with 1 mM levamisole overnight at room temperature . The color reaction was stopped after observation of sufficient development of blue precipitate by washing with PBST . The slides were then counterstained with Fastred ( Sigma Aldrich ) , mounted and coverslipped . MRC5 cells were grown on a Round Glass Coverslips Ø 16 mm ( thermo scientific ) placed inside a 12 Multiwell Plate . Coverslips slides were washed in phosphate-buffered saline and fixed in 4% paraformaldehyde for 15 min , cells were then permeated using 0 . 1% Triton X-102 ( Agilent Technologies ) for 10 min . and blocked with PBS solution containing BSA ( 3% ) for 30 min . Incubation with primary antibodies was performed in a bloking solution BSA ( 1% ) at 37°C for 1 h at the following dilutions; α-SMA ( 1∶1000 ) , CAV1 ( 1∶50 ) , . After three washes with PBS , cells were incubated with secondary Alexa Fluor 488 goat anti-Mouse IgG ( Invitrogen ) ( 1∶500 ) , Alexa Fluor 647 goat anti-rabbit IgG ( Invitrogen ) ( 1∶500 ) and Alexa Fluor 647 Phalloidin ( A22287 - Life technologies ) ( 1Unit/slide ) . Fourty five min later , Coverslips slides were fixed on microscope slides using ProLong Gold Antifade Reagent with DAPI ( Invitrogen ) . Fluorescence was viewed with an FV10i Olympus confocal scanning microscope . MRC5 cells ( 150 , 000/well ) were seeded in duplicate in DMEM supplemented with 10% FBS on 60-mm cell culture dishes . Cells were serum starved the next day and transfected with pre-miR-199a-5p at 10 nM . Cell proliferation was assessed 48 h after transfection by flow cytometry using the click-iT EdU cell proliferation assay ( Invitrogen ) according to the manufacturer's instructions . hFL1 cells were seeded on Type-I collagen coated 12-well plates and transfected as described above . Forty eight hours after transfection , confluent cells were ( FCS ) starved 3 h before adding 10 ng/ml TGFβ and wounded using a pipet tip . The in vitro wound-healing process was then recorded by videomicroscopy for 24 h from then scratching on an Axiovert 200 M inverted microscope ( Carl Zeiss ) equipped with 37°C and 5% CO2 regulated insert ( Pecon GmbH ) . Brightfied images were taken each 30 min through a 10× phase contrast objective with a CoolSNAPHQ CCD Camera managed by Metamorph Software ( Roper Scientific ) . The motility of the cells was assessed by evaluating the repaired area percentage using ImageJ sotware . Invasion of MRC5 fibroblast overexpressing miR-199a-5p was assessed using commercially available 24-well BioCoat Matrigel Invasion Chamber ( BD Biosciences ) . In brief , pulmonary fibroblasts were transfected either with pre-miR-199a-5p or negative control as described above . Twenty four hours after transfection , cells were harvested with trypsin-EDTA , centrifuged , and resuspended in DMEM medium . Cell suspensions ( 1×105 cells/well ) were added to the upper chamber . Bottom wells of the chamber were filled with DMEM medium containing 10% FBS as chemoattractant , whereas the upper chamber was filled with DMEM only . After incubation for 48 h at 37°C , the non-invading cells on the top of the membrane were removed with a cotton swab . Membrane containing invading-cells were fixed with methanol , washed three times with PBS and mounted with DAPI hard set ( Vector Laboratories ) onto glass slides for fluorescent microscopy . Results are given as mean±S . E . M . Statistical analyses were performed by using Student's t-test as provided by Microsoft Excel . | Fibrosis is the final common pathway in virtually all forms of chronic organ failure , including lung , liver , and kidney , and is a leading cause of morbidity and mortality worldwide . Fibrosis results from the excessive activity of fibroblasts , in particular a differentiated form known as myofibroblast that is responsible for the excessive and persistent accumulation of scar tissue and ultimately organ failure . Idiopathic Lung Fibrosis ( IPF ) is a chronic and often rapidly fatal pulmonary disorder of unknown origin characterized by fibrosis of the supporting framework ( interstitium ) of the lungs . Given the poor prognosis of IPF patients , new insights into the biology of ( myo ) fibroblasts is of major interest to develop new therapeutics aimed at reducing ( myo ) fibroblast activity to slow or even reverse disease progression , thereby preserving organ function and prolonging life . MicroRNAs ( miRNAs ) , a class of non-coding RNA recently identified , are associated with normal cellular processes; and deregulation of miRNAs plays a causative role in a vast array of complex diseases . In this study , we identified a particular miRNA: miR-199a-5p that governs lung fibroblast activation and ultimately lung fibrosis . Overall we showed that miR-199a-5p is a major regulator of fibrosis with strong therapeutic potency to treat fibroproliferative diseases such as IPF . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"medicine",
"histology",
"functional",
"genomics",
"interstitial",
"lung",
"diseases",
"gene",
"regulation",
"genetics",
"molecular",
"genetics",
"biology",
"genomics",
"genetics",
"and",
"genomics",
"respiratory",
"medicine",
"pulmonology"
] | 2013 | miR-199a-5p Is Upregulated during Fibrogenic Response to Tissue Injury and Mediates TGFbeta-Induced Lung Fibroblast Activation by Targeting Caveolin-1 |
Histoplasmosis is an endemic fungal infection in French Guiana . It is the most common AIDS-defining illness and the leading cause of AIDS-related deaths . Diagnosis is difficult , but in the past 2 decades , it has improved in this French overseas territory which offers an interesting model of Amazonian pathogen ecology . The objectives of the present study were to describe the temporal trends of incidence and mortality indicators for HIV-associated histoplasmosis in French Guiana . A retrospective study was conducted to describe early mortality rates observed in persons diagnosed with incident cases of HIV-associated Histoplasma capsulatum var . capsulatum histoplasmosis admitted in one of the three main hospitals in French Guiana between 1992 and 2011 . Early mortality was defined by death occurring within 30 days after antifungal treatment initiation . Data were collected on standardized case report forms and analysed using standard statistical methods . There were 124 deaths ( 45 . 3% ) and 46 early deaths ( 16 . 8% ) among 274 patients . Three time periods of particular interest were identified: 1992–1997 , 1998–2004 and 2005–2011 . The two main temporal trends were: the proportion of early deaths among annual incident histoplasmosis cases significantly declined four fold ( χ2 , p<0 . 0001 ) and the number of annual incident histoplasmosis cases increased three fold between 1992–1997 and 1998–2004 , and subsequently stabilized . From an occasional exotic diagnosis , AIDS-related histoplasmosis became the top AIDS-defining event in French Guiana . This was accompanied by a spectacular decrease of early mortality related to histoplasmosis , consistent with North American reference center mortality rates . The present example testifies that rapid progress could be at reach if awareness increases and leads to clinical and laboratory capacity building in order to diagnose and treat this curable disease .
French Guiana is a French overseas territory , located in the North-Eastern part of South America . The Human Immunodeficiency Virus ( HIV ) epidemic there is the most preoccupying among French territories [1] . During the Highly Active AntiRetroviral Therapy ( HAART ) era , disseminated histoplasmosis has remained the most common Acquired Immunodeficiency Syndrome ( AIDS ) defining illness with an incidence of 15 . 4/1000 person-years in HIV-infected patients [2] . In immunocompetent patients , Histoplasma capsulatum var . capsulatum infection is typically asymptomatic or pauci-symptomatic and spontaneous resolution is the rule in the great majority of cases [3] . On the contrary , in HIV-infected patients it presents mostly as a disseminated infection . With the worsening of the immunosuppression , the disease progression is often rapid and always fatal in the absence of treatment [4] . Thus , different studies have observed up to 39% of deaths following diagnosis in endemic areas , where it is supposedly well known , and 58% in non endemic areas , where it is perhaps less known [5] , [6] . In endemic areas , although there are different outcome measures and inclusion criteria , the death rates observed in AIDS-associated histoplasmosis differ between the USA ( 12–23% ) and South America ( 19–39% ) [6] . Hypotheses advanced to explain these differences are a delayed recognition due to the lack of awareness of physicians , a delayed diagnosis due to the lack of diagnostic facilities and the late presentation of HIV-infected patients in resource limited settings [6] , [7] , [8] . Delayed treatment due to the unavailability of the most effective therapy in severe cases , the impossibility of monitoring drug concentrations and/or drug-drug interactions with antituberculosis treatments are other possible explanations [6] . In French Guiana , disseminated histoplasmosis has also been the leading cause of death among HIV-infected patients [9] . Despite HIV care and treatment standards close to those in Mainland France , the mortality rate of AIDS-associated histoplasmosis remains high in the HAART era ( 30 . 7% at 6 months and 17 . 5 at 1 month ) , whereas in Mainland France , a non-endemic area , this mortality rate was divided by two [10] , [11] . The objective of this study was to describe the temporal trends of incidence and mortality indicators for AIDS-associated histoplasmosis in French Guiana . This knowledge is important to guide and improve AIDS-associated histoplasmosis diagnosis , care and treatment , and to illustrate that awareness and standard practices in mycology can dramatically change prognosis .
Since 1992 , an anonymized database compiles retrospectively and continuously Histoplasma capsulatum var . capsulatum histoplasmosis confirmed incident cases diagnosed in HIV-infected patients according to the case definition of the European Organization for Research and Treatment of Cancer/Invasive Fungal Infections Cooperative Group and the National Institute of Allergy and Infectious Diseases Mycoses Study Group ( EORTC/MSG ) Consensus Group [12] . The revised EORTC/MSG criteria defining a proven case of histoplasmosis are: recovery in culture from a specimen obtained from the affected site or from blood; and/or histopathologic or direct microscopic demonstration of appropriate morphologic forms with a truly distinctive appearance characteristic such as intracellular yeasts forms in a phagocyte in a peripheral blood smear or in tissue macrophages . By contrast , molecular methods of detecting fungi in clinical specimens , such as Polymerase Chain Reaction ( PCR ) , were not included in the classifications of “proven , ” “probable , ” and “possible” invasive fungal disease ( IFD ) definitions because there is as yet no standard , and none of the techniques has been clinically validated . All HIV-infected patients hospitalized or seen in the outpatient department before admission , suspicious for histoplasmosis and receiving antifungal therapy in one of the three main hospitals of French Guiana ( the Centre Hospitalier de Cayenne ( CHC ) , the Centre Hospitalier Médico-Chirurgical de Kourou ( CMCK ) and the Centre Hospitalier de l'Ouest Guyanais in Saint Laurent du Maroni ( CHOG ) , were identified and checked for a confirmed diagnosis of histoplasmosis in all laboratories where biological samples were sent . Then , they were finally enrolled according to the following inclusion criteria: age >18 years , admission in one of the three hospitals ( the inclusion date corresponding to the date of antifungal treatment initiation ) , confirmed HIV infection ( by Western blot ) , confirmed incident histoplasmosis infection ( EORTC/MSG criteria ) , and baseline blood screening within 7 days prior to antifungal therapy initiation . Non inclusion criteria were: histoplasmosis relapse or diagnosis of histoplasmosis relying only on Histoplasma Polymerase Chain Reaction ( PCR ) . Data were collected on a standardized form and included sociodemographic , clinical , biologic , radiologic , therapeutic and survival information . These data were then entered in an anonymized database . Ethical approval was obtained for the database and related studies ( IRB0000388 , FWA00005831 ) . A descriptive study of the patients included in this database until April 2007 was published elsewhere [10] . An observational , retrospective and multicentric study was conducted from 01/01/1992 to 09/30/2011 , using the French Guiana HIV-Histoplasmosis database described above . In this study , the primary endpoint was the vital status on day 30 following antifungal therapy initiation . Patients lost to follow up within 30 days following antifungal therapy initiation , or deceased with an unknown date of death , or presenting a relapse of histoplasmosis were excluded from the analysis . This early death criterion appeared as a good compromise to attribute mortality to the histoplasmosis infectious episode under consideration , in a context of severe immunosuppression favouring multiple opportunistic pathogens , ensuring simplicity and reproducibility of the study . The statistical analysis was performed using STATA 10 . 0 ( College Station , Texas , USA ) ( 38 ) . Descriptive analysis used proportions , medians and trend χ2 test .
There were 278 patients with AIDS-associated histoplasmosis . Four cases were excluded before the analysis ( 3 because they were lost to follow up and one because of an unknown date of death ) . Their socio-demographic characteristics and median CD4 count did not differ from the 274 patients finally selected in this study ( data not shown ) . Among the 274 patients selected for whom the vital status at 30 days after antifungal therapy initiation was known , there were 124 deaths ( 45 . 3% ) . The median time to death was 110 days ( Interquartile Range [IQR] = 13–481 ) and the median age at the time of death was 39 years ( IQR = 33–47 ) . Early death occurred in 46 patients ( 16 . 8% ) with a median survival time of 7 days ( IQR = 3–16 ) after antifungal treatment initiation . The median age at the time of early death was 37 years ( IQR = 32–47 ) . Figure 1 shows that the proportion of deaths occurring the same year as the diagnosis of incident histoplasmosis cases remained stable around 5 deaths per year until 2005/2006 and then stabilized around 3 deaths per year . Among these deaths cases , almost half were early deaths until 2004 . From 2005 onwards there was a notable decline of early deaths along with the overall decline of mortality . In addition , starting in 1998 , the number of histoplasmosis cases diagnoses increased , and subsequently the number of incident cases oscillated between 14 and 22 cases per year . Data were incomplete for 2011 , the study considering cases only until 09/30/2011 . Thus , three time periods of particular interest have been identified: 1992–1997 , 1998–2004 and 2005–2011 . Figure 2 summarizes the two main temporal trends observed in Figure 1 . First , the proportion of early deaths among annual incident histoplasmosis cases was significantly divided four fold ( χ2 , p<0 . 0001 ) . Second , the number of annual incident histoplasmosis cases increased three fold between 1992–1997 and 1998–2004 , and subsequently stabilized at the same level . Table 1 showed that early deaths associated with histoplasmosis occurred mainly in men , late presenters with HIV infection ( CD4 count <50/mm3 ) among whom 10% were on HAART on admission . The incident histoplasmosis cases were mainly disseminated and often recognized as the first AIDS-defining illness in the course of HIV infection . Fungal culture and direct examination were the main methods used for the diagnosis of histoplasmosis cases . The Real Time Polymerase Chain Reaction ( RT-PCR ) detection method for Histoplasma only became available during the 2005–2011 period . Amphotericin B and itraconazole were the first line antifungal regimen used to treat these patients . During the study period , liposomal amphotericin B and itraconazole became the standard antifungal regimen over deoxycholate amphotericin B and fluconazole , respectively .
This study described 19 years of experience in French Guiana . Three periods of interest and two main trends could be observed from 1998 onwards: the spectacular decrease of early deaths among incident histoplasmosis cases , and a simultaneous marked increase of the annual incidence of histoplasmosis cases . Whereas , during the same period , HIV prevalence in pregnant women was quite stable >1% since the 1990's: 0 . 8%–1 . 4% between 1992–1997 , 1 . 2%–1 . 4% between 1998–2004 and 1 . 0%–1 . 2% between 2005–2011 [1] , [13] . The increased number of annual histoplasmosis cases can be attributed to the development of medical mycology skills in hospitals laboratories , notably a reference university laboratory specialized in parasitology-mycology established since 1997 in Cayenne Hospital . By the same time , highly active antiretroviral therapy was introduced , which could have led to more patent cases of histoplasmosis due to the immune reconstitution inflammatory syndrome [14] . In addition , a PCR diagnostic method became available for histoplasmosis in 2006 [15] . Unfortunately , urinary antigen detection for histoplasmosis is still unavailable in French Guiana . The sharp decline of the proportion of early deaths can be attributed to the improvement of the diagnostic capacity along with the improvement of the clinical management of HIV-infected patients following French recommendations [16] . Thus , French Guiana reached HIV-virological suppression levels comparable to those in Mainland France by 2004 . In addition , this trend can also be attributed to the improvement of the clinical management of AIDS-related disseminated histoplasmosis cases . The accurate recognition of severe cases and the supply of liposomal amphotericin B since 1998 , an effective and less nephrotoxic treatment recommended for severe disseminated histoplasmosis cases , were two important factors behind the progress . This study had limitations . Data were collected retrospectively , which might have led to selection biases . Determining retrospectively if death was related to AIDS-associated histoplasmosis incident cases under study is challenging , considering the high percentage of concomitant opportunistic infections . Thus , we chose early death as the primary outcome because we thought that retrospectively it was the simplest and most reproducible indicator of histoplasmosis AIDS-related deaths . Despite its limitations , this study showed that capacity building both in laboratory and clinical practice , effective drug availability both for HIV and histoplasmosis infections , and an effective bench to bed collaboration between actors progressively helped in reducing the burden of overall deaths and early deaths . Mortality indicators are now consistent with those described in North America , where the most effective and non invasive histoplasmosis diagnostic method is available . To further reduce early mortality , reducing diagnostic delays and antifungal therapy initiation is still a major objective . To reach it , a diagnostic method that meets the World Healh Organization's A . S . S . U . R . E . D . ( Affordable , Sensitive , Specific , User-friendly , Rapid/Robust , Equipment-free and Delivered ) should be developed . Although our results may seem parochial , they illustrate the rapid progress that took place within a decade . The increased awareness of clinicians , who became more aggressive in their investigations , and the increased laboratory capacity led to find and treat a disease that was present but probably not identified and not treated in time . Thus , histoplasmosis , previously known as a mild disease in immunocompetent individuals , became a public health problem in HIV-infected patients , known by almost all health practitioners in French Guiana . By dealing with the mycology diagnostic tool box limitations and starting prompt presumptive antifungal treatment in HIV-infected patients it was possible to reduce early deaths considerably . The historical 40% of early deaths observed in French Guiana , where histoplasmosis was known , plausibly reflects a low estimate of what happens in the Amazon region and probably beyond , where histoplasmosis is endemic but probably still widely misdiagnosed for tuberculosis and/or neglected [17] . Although cost effective strategies to prevent the disease and very effective diagnostic methods have been developed and are well known by scattered medical teams in Latin America [18] , this knowledge does not percolate to too many HIV care units and hospital laboratories [19] . The present example testifies that rapid progress could be at reach if awareness increased and led to implement clinical and laboratory capacity building in order to diagnose and treat this curable disease before it is too late . | Histoplasmosis is an endemic fungal infection in French Guiana . It is the most common AIDS-defining illness and the leading cause of AIDS-related deaths . Diagnosis is difficult , but in the past 2 decades , it has improved . The objectives of the present study were to describe the temporal trends of incidence and mortality indicators for HIV-associated histoplasmosis in French Guiana . A retrospective study was conducted to describe early mortality rates observed in persons diagnosed with incident cases of HIV-associated histoplasmosis admitted in one of the three main hospitals of French Guiana between 1992 and 2011 . Early mortality was defined by death occurring within 30 days after antifungal treatment initiation . Data were collected on standardized case report forms and analysed using standard statistical methods . Among 274 patients there were 46 early deaths ( 16 . 8% ) . The two main temporal trends were: the proportion of early deaths significantly divided four fold and the number of annual incident histoplasmosis cases increased three fold . The present example testifies that rapid progress could be at reach if awareness increases and leads to clinical and laboratory capacity building in order to diagnose and treat this curable disease . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"infectious",
"diseases",
"medicine",
"and",
"health",
"sciences",
"sexually",
"transmitted",
"diseases",
"aids",
"histoplasmosis",
"neglected",
"tropical",
"diseases",
"fungal",
"diseases",
"viral",
"diseases",
"infectious",
"disease",
"control",
"tropical",
"diseases"
] | 2014 | HIV-Associated Histoplasmosis Early Mortality and Incidence Trends: From Neglect to Priority |
Borrelia persica , a bacterium transmitted by the soft tick Ornithodoros tholozani , causes tick-borne relapsing fever in humans in the Middle East , Central Asia and the Indian peninsula . Immunocompetent C3H/HeOuJ mice were infected intradermally with B . persica at varying doses: 1 x 106 , 1 x 104 , 1 x 102 and 4 x 100 spirochetes/mouse . Subsequently , blood samples were collected and screened for the presence of B . persica DNA . Spirochetes were detected in all mice infected with 1 x 106 , 1 x 104 and 1 x 102 borrelia by real-time PCR targeting the flaB gene of the bacterium . Spirochetemia developed with a one- to two-day delay when 1 x 104 and 1 x 102 borrelia were inoculated . Mice injected with only four organisms were negative in all tests . No clinical signs were observed when infected mice were compared to negative control animals . Organs ( heart , spleen , urinary bladder , tarsal joint , skin and brain ) were tested for B . persica-specific DNA and cultured for the detection of viable spirochetes . Compiled data show that the target organs of B . persica infections are the brain and the skin . A newly developed serological two-tiered test system ( ELISA and western blot ) for the detection of murine IgM , IgG and IgA antibody titers against B . persica showed a vigorous antibody response of the mice during infection . In conclusion , the infection model described here for B . persica is a platform for in vivo studies to decipher the so far unexplored survival strategies of this Borrelia species .
Spirochetes of the genus Borrelia ( B . ) are vector-borne , spiral-shaped bacteria that can be divided into two functional groups [1] . One large group of spirochetes belongs to the B . burgdorferi sensu lato complex ( e . g . B . burgdorferi sensu stricto , B . afzelii , B . garinii , B . bavariensis ) . Lyme disease borreliae are transmitted by hard-shelled Ixodes ticks [2] . The second group includes the relapsing fever ( RF ) borreliae which spread primarily via soft ticks with an exception of B . recurrentis that is transmitted by the body louse ( Pediculus humanus ) . Among others , B . hermsii , B . duttonii and also B . persica are tick-borne RF borreliae which induce tick-borne relapsing fever ( TBRF ) ( reviewed in [3] ) . B . persica is transmitted by the soft tick Ornithodoros tholozani during blood meals [4] . These ticks are prevalent in areas such as the Middle East , Central Asia and the Indian peninsula and feed on humans as well as on animals ( reviewed in [5] ) . Moreover , TBRF can occur in non-endemic countries due to travel of infected people [6 , 7] . Clinically , the disease manifests with fever attacks in human patients related to high numbers of spirochetes in the blood circulation during fever episodes [8–10] and non-specific clinical signs such as chills , headache , nausea , vomiting , sweating , abdominal pain , arthralgia , cough and photophobia which may occur [9] . Rodhain reviewed as early as 1976 [11] that a high level of experimental pathogenicity of B . persica can be perceived in guinea pigs , hedgehogs and rabbits whereas lower levels seem to occur in monkeys , adult white mice and rats . So far primarily guinea pigs have been used to multiply B . persica [12–14] and just recently it was possible to maintain B . persica in vitro [14 , 15] . However , for other RF borrelia species the mouse is usually considered to be the appropriate animal model [16–19] and Babudieri investigated relapsing fever in Jordan by injecting blood of diseased patients into mice in order to confirm TBRF spirochetosis [20] . The mice used in the experiment tested positive two to five days after injection . Furthermore , Babudieri studied the infection rate of captured Ornithodoros tholozani ticks . Squashed ticks were inoculated into mice , but infection was not initiated , while a very low infection rate was obtained when the ticks were allowed to feed directly on these animals . Spirochetes were not present constantly and uniformly in the mice’s blood . In addition , the author mentioned that the spirochetes survived in the mice’s brains . In 2006 , Assous et al . inoculated intraperitoneally blood of TBRF patients from Israel into ICR mice and detected spirochetes in the mice’s blood samples on day four as well as on day six of the experiment [21] . Since comprehensive data for B . persica in mice were not available , we aimed to establish and characterize in detail an infection model for B . persica in immunocompetent mice . Therefore , we infected intradermally 44 C3H/HeOuJ mice with B . persica strain LMU-C01 . In order to gain insight into the infection with these TBRF spirochetes , we investigated ( a ) whether this laboratory strain of B . persica is able to establish an infection in immunocompetent mice; ( b ) whether the mice develop clinical signs; ( c ) when and in which quantity the spirochetes appear in the blood circulation; ( d ) confirm that B . persica disseminates into organs and investigate the histopathological changes; ( e ) characterize the mice’s immune response during infection; and ( f ) define the minimal dose necessary to infect animals . After compilation of all data , we came to the conclusion that the infection model described here is a reliable tool that can be used for further research studies .
For this study , 100 μl of thawed B . persica passages ( strain LMU-C01 , isolated from a cat in Israel; passage 2 , 3 . 9 x 106 organisms/ml ) were cultivated in Pettenkofer/LMU Bp medium as described previously [15] . Cultures were incubated for five days and viable bacteria were counted with a Petroff-Hausser counting chamber ( Hausser Scientific , Horsham , Pennsylvania , USA ) . Bacteria suspensions were adjusted to the required cell concentration by dilution of cultures with plain medium . In total , 54 six- to seven-week-old female C3H/HeOuJ mice ( Charles River Wiga Deutschland GmbH , Sulzfeld , Germany ) were kept in individually ventilated cages ( ISOcage N System; Tecniplast Deutschland GmbH , Hohenpeißenberg , Germany ) at the animal facility of the Institute for Infectious Diseases and Zoonoses , Ludwig-Maximilians-Universität ( Munich , Germany ) . Animals were manipulated in laminar flow systems in order to sustain specific-pathogen-free conditions . The health status of all mice and the body temperature , which was measured with a subcutaneous transponder ( IPTT-300 Temperature Transponders; Plexx B . V . , PW Elst , Netherlands ) , were recorded twice a day . Initially , 20 mice were exposed to 1 x 106 B . persica spirochetes in 100 μl medium by intradermal injection into the shaven back . In addition , four animals were injected with 100 μl medium alone and served as negative controls . The injection volume was divided into small portions ( 10 x 10 μl ) , placed close to each other into the skin ( ~ 4 cm2 area ) . For the dose finding study , eight mice per group were injected with B . persica suspensions with varying concentrations . Group #1: 1 x 104 spirochetes per mouse; group #2: 1 x 102 spirochetes per mouse; group #3: 4 x 100 spirochetes per mouse . Two additional animals in each group served as negative and infection/transmission controls . To study the kinetics of bacteremia and the development of specific antibodies post infection , blood samples were collected at preassigned intervals . Two drops of blood were collected in a Microvette 100 K3E ( preparation K3EDTA; Sarstedt AG & Co . , Nümbrecht , Germany ) by facial bleeding after cutting the skin with a 4-mm Goldenrod Animal Lancet ( Braintree Scientific , BioMedical Instruments , Zöllnitz , Germany ) . The bleeding scheme was as follows: during the first two weeks each mouse was bled every second day . However , in order to collect data for each single day of the first 14-day interval , the group was divided into two equal subgroups and these subgroups were bled according to alternating schedules . After day 14 , all animals were bled together once a week until the final days 49/50 . The bleeding scheme for the dose finding study was: during the first 20 days , each animal was bled every second day . Subgroups were formed and bled according to alternating schedules to obtain blood samples for each experimental day . After day 20 , blood samples were collected every second day up to the final days 30/31/32 . Alternating schedules were applied to the subgroups ( each mouse was bled every fourth day ) . For DNA-extraction , 5 μl from each blood sample were transferred into a 1 . 5-ml safe-lock tube ( Eppendorf Vertrieb Deutschland GmbH , Wesseling-Berzdorf , Germany ) and frozen at -30°C until used . Surplus blood samples of the regular blood collection from animals that had received 1 x 106 B . persica organisms were pooled subgroup-specific in another 1 . 5-ml safe-lock tube for plasma production . After euthanasia , a final blood sample of each mouse was collected in a micro tube ( 1 . 1ml Z-Gel; Sarstedt AG & Co . ) for serum production . Plasma and serum preparation were done by a two-time centrifugation ( Centrifuge 5430 R V 1 . 1 , rotor FA-45-30-11; Eppendorf Vertrieb Deutschland GmbH ) at 350 x g for 10 min at 24°C . The supernatants were collected in a 1 . 5-ml safe-lock tube and were frozen at -30°C until serological analyses were performed . AS3000 Maxwell 16 MDx Instrument and the Maxwell 16 LEV Blood DNA Kit ( Promega GmbH , Mannheim , Germany ) were used for DNA extraction from blood and tissue samples . In the case of blood: 5 μl thawed blood , 300 μl sterile phosphate-buffered saline ( PBS ) , 300 μl lysis buffer and 30 μl Proteinase K were mixed . The following steps were done according to the manufacturer’s technical manual # TM333 ( Maxwell 16 LEV Blood DNA Kit and Maxwell 16 Buccal Swab LEV DNA Purification Kit Technical Manual; Promega GmbH ) . DNA was eluted in 60 μl elution buffer and frozen at -30°C . In the case of tissue: 200 μl of incubation buffer ( Promega GmbH ) were filled in a 1 . 5-ml safe-lock tube containing thawed tissue ( weight less than 50 mg ) . Then , 200 μl lysis buffer and 30 μl Proteinase K were added and the tissue sample was squeezed and disrupted with a micro pestle ( Faust Lab Science GmbH , Klettgau , Germany ) . Samples were incubated in a ThermoMixer comfort 5355 V 2 . 0 ( Eppendorf Vertrieb Deutschland GmbH ) at 56°C and 500 rpm overnight . Additional 200 μl of lysis buffer were added and DNA was extracted with the Maxwell 16 MDx Instrument . DNA was eluted and frozen as written above . B . persica DNA was detected with a real-time quantitative PCR ( qPCR ) assay in a Mx3005P qPCR System ( Agilent Technologies Sales & Services GmbH & Co . KG , Böblingen , Germany ) . The primers and the probe were designed according to the flaB target gene of B . persica using the software Primer3Plus ( Free Software Foundation , Inc . , Boston , Massachusetts , USA; http://primer3plus . com; [22] ) . Synthesis of following sequences was carried out by Eurofins Genomics ( Ebersberg , Germany ) : Bp_flaB_fw 5’-GAG GGT GCT CAA CAA GCA A-3’ , Bp_flaB_probe 5’-FAM-AAA TCA GGA AGG AGT ACA ACC AGC AGC A-3’-TAM and Bp_flaB_re 5’-CAA CAG CAG TTG TAA CAT TAA CTG G-3’ . The expected amplicon size was 106 base pairs . Real-time PCR was carried out in 96 Multiply PCR plate natural ( Sarstedt AG & Co . ) containing 1 . 2 μl of each primer ( final concentration 600 nM ) , 0 . 8 μl of the probe ( final concentration 200 nM ) , 10 μl GoTaq Probe qPCR Master Mix ( 2 x; Promega GmbH; final concentration 1 x , added CXR reference dye following the manufacturer’s protocol ) and 2 . 5 μl target DNA solution . The reaction volume was 20 μl in total and was pipetted in duplicate for each DNA sample . The amplification program was as follows: initial activation at 95°C for 5 min , 40 cycles of 95°C for 15 s and 60°C for 60 s and a final step at 25°C for 15 s . In each qPCR run a positive control ( B . persica strain LMU-C01 , P3 ) , no template control ( NTC , 2 . 5 μl nuclease-free water; Promega GmbH ) and samples for calibration of the standard curve ( serial dilution of B . persica DNA , P4 ) were included . According to the standard curve ( considering slope , efficiency and R-squared value ) , the absolute spirochete number per ml mouse blood was calculated using the MxPro QPCR Software version 4 . 10 ( Agilent Technologies Sales & Services GmbH & Co . KG ) based on threshold cycles ( Ct ) . Graphs were constructed with the OriginPro 9 . 1 Software ( Additive GmbH , Friedrichsdorf , Germany ) . As regards tissue , additional to DNA of B . persica flaB gene mouse-specific glyceraldehyde-3-phosphate dehydrogenase ( GAPDH; TaqMan Gene Expression Assay , Mm99999915_g1 , VIC dye-labeled MGB probe , 20 x; Applied Biosystems , Thermo Fisher Scientific GmbH , Ulm , Germany ) was detected to control the DNA content in the tissue sample . TaqMan Gene Expression Assay was used according to the manufacturer’s recommendations . The reactions mix ( total volume 20 μl ) contained 10 μl GoTaq Probe qPCR Master Mix ( Promega GmbH; final concentration 1 x , added CXR reference dye ) , 1 μl of the TaqMan Gene Expression Assay ( final concentration 1 x ) and 2 . 5 μl DNA solution . PCR conditions were as described above , with the exception of the initial activation step that was separated into two steps: 50°C for 2 min followed by 95°C for 10 min . At the end of the infection study with 1 x 106 borrelia per mouse , animals were euthanized at days 49/50 post infection . Tissue processing was carried out as described previously [23] . Heart , spleen , urinary bladder , left tarsal joint , skin from infection areal and brain were collected from each mouse and divided into two parts ( brain into three parts ) . One part was put in a 1 . 5-ml safe-lock tube and frozen at -30°C for DNA-extraction . The other part was transferred into a second 1 . 5-ml safe-lock tube filled with 200 μl of Pettenkofer/LMU Bp medium . Subsequently , the tissue was squeezed and disrupted with a micro pestle and the suspension was transferred into a 12-ml tube ( Centrifuge Tube 12; TPP , Faust Lab Science GmbH ) filled with 10 ml Pettenkofer/LMU Bp medium . The organ cultures were incubated at 37°C in humidified air for three weeks . Observation for viable mobile spirochetes was performed weekly using a dark-field microscope . The third part of the brain , right kidney and right tarsal joint were transferred into a 50-ml centrifuge tube ( 114x28mm , PP; Sarstedt AG & Co . ) filled with 20 ml of 4% formalin and were stored at room temperature until histopathology analyses were carried out . For the dose finding study the skin , brain , right kidney and right tarsal joint were collected on final days 30/31/32 and prepared as described above . Brain parts , right kidneys and right tarsal joints ( in 4% formalin ) from five infected mice ( infection dose 1 x 106 B . persica/mouse ) and one control animal were used for the histopathological evaluations . Parts of the brains and kidneys were embedded in paraffin and were cut into 2–3 μl thin slices . The other parts of brains and kidneys as well as tarsal joints were embedded in plastic and sectioned into 1 μl thin slices . After staining with hematoxylin and eosin ( HE ) as well as Giemsa , observations for histopathological changes were carried out under a bright-field microscope . A low-passaged culture of B . persica ( strain LMU-C01 ) was used for antigen production . The purified bacterial lysate was utilized to detect mouse antibodies in the serological two-tiered test system ( ELISA and western blot ) . Spirochetes were first cultured as described elsewhere [15] . When bacteria reached the late exponential phase ( after five days ) , 150 μl were transferred into each of two 12-ml tubes ( Centrifuge Tube 12; TPP , Faust Lab Science GmbH ) containing 6 ml medium and were further incubated for three days . These 6-ml bacteria suspensions were transferred to a sterile glass bottle containing 1 l medium and incubated until late exponential phase of growth ( five days of incubation ) . Antigen preparation was done via ultrasound disruption according to Töpfer et al . [24] and the centrifuged supernatant of the whole cell lysate was stored at -80°C until used . Determination of protein concentration and quality control of the antigen solution was carried out as described previously [24] . The microdilution plates ( Nunc-Immuno Microwell Maxisorp C96; Thermo Scientific , VWR International GmbH , Ismaning , Germany ) were coated with whole cell antigen lysate of B . persica at a concentration of 0 . 2 μg per well as described by Barth et al . [25] . Detection of specific antibodies against B . persica was done with a computer-assisted , kinetic-based ELISA after Shin et al . [26] . Serum and pooled plasma samples were diluted 1:100 in sample buffer containing PBS , 0 . 05% Tween 20 ( neoLab Migge Laborbedarf-Vertriebs GmbH , Heidelberg , Germany ) and 2% non-fat dry milk ( Merck KGaA , Darmstadt , Germany ) . Four control serum samples were added in each run . Peroxidase-conjugated goat IgG fraction to mouse immunoglobulins ( IgG , IgA , IgM; MP Biomedicals , LLC , Heidelberg , Germany ) were diluted 1:4 , 000 in sample buffer and used as the secondary antibody . As a final step , substrate ( TMB Microwell Peroxidase Substrate Kit; KPL , medac GmbH , Wedel , Germany ) was added and after 1 min 45 s the extinction of each well was read five times in 35-s intervals at 650 nm in a SpectraMax Plus 384 Microplate Reader ( Molecular Devices ( UK ) Ltd , Wokingham , United Kingdom ) . Results were calculated with the SoftMax Pro software 5 . 3 ( Molecular Devices ( UK ) Ltd ) . To standardize the sample evaluation and for the comparability of the plates of each run , the results of the samples were adjusted to the evaluated values of the control samples . Graphs were constructed with the OriginPro 9 . 1 Software ( Additive GmbH ) . For antigen preparation , three parts of antigen were mixed with one part of reducing sample buffer ( Roti-Load 1; Carl Roth GmbH & Co . KG , Karlsruhe , Germany ) and heated for 10 min at 90°C in a ThermoMixer comfort 5355 V 2 . 0 ( Eppendorf Vertrieb Deutschland GmbH ) . The diluted antigen was loaded into a precast gel ( 4–15% Mini-PROTEAN TGX Stain-Free Precast Gels , IPG well comb , 86 x 67 mm ( W x L ) ; Bio-Rad Laboratories GmbH , Munich , Germany ) and a protein weight ladder ( Precision Plus Protein WesternC Standards; Bio-Rad Laboratories GmbH ) was included separated from each other with a 5-mm wide polytetrafluoroethylene ( PTFE ) stick . Gel electrophoresis was performed with 1:10 diluted running buffer ( 10x Tris/Glycine/SDS Buffer; Bio-Rad Laboratories GmbH ) in a Mini-PROTEAN Tetra cell ( Bio-Rad Laboratories GmbH ) at 250 V for 22 min . Western blot and immunodetection were done according to the Protein Blotting Guide ( Bulletin #2895; Bio-Rad Laboratories GmbH ) and a house-intern protocol as outlined below . Buffers and solutions were produced following recipes of the Protein Blotting Guide . Blotting of proteins onto a nitrocellulose membrane ( MemBlot CN—Rolle , 0 . 45 μm , 10 x 7 . 5 cm; membraPure , Bodenheim , Germany ) was carried out at 30 V for 960 min using a Mini Trans-Blot module ( Bio-Rad Laboratories GmbH ) in Towbin Buffer . The membrane was washed with tris-buffered saline ( TBS , pH = 7 . 5 ) for 7 min and blocked for 1 h at room temperature in 5% non-fat milk-TBS . Subsequently , the membrane was washed with TTBS twice for 7 min ( 0 . 05% Tween 20 in TBS ) and subsequently cut into strips ( 3–4 mm wide ) . Serum and plasma samples were diluted 1:100 in 5% non-fat milk-TTBS and incubated with the membrane strips for 1 h at room temperature hhhhhhh . Protein standard strips were incubated with plain 5% non-fat milk-TTBS . After washing ( four times , 7 min , in TTBS ) , the strips were incubated with 1:1 , 000 diluted detection antibody in TTBS ( peroxidase-conjugated goat IgG fraction to mouse immunoglobulins IgG , IgA , IgM; MP Biomedicals , LLC ) , and the strips with the protein standard were incubated with Streptactin solution ( Precision Protein StrepTactin-HRP Conjugate; Bio-Rad Laboratories GmbH ) for 1 h at room temperature , respectively . Strips were washed four times with TTBS for 7 min . After a final wash step with TBS for 1 min , color development was achieved by adding substrate ( Opti-4CN Substrate Kit; Bio-Rad Laboratories GmbH ) and stopped after 4 min by washing in distilled water . Images were taken with the CemiDoc MP System and Image Lab Software Version 5 . 0 ( Bio-Rad Laboratories GmbH ) . Mouse experiments were carried out according to the guidelines approved by the Animal Welfare Committee of the Sachgebiet 54 , Regierung von Oberbayern ( Munich , Germany ) . The animal care and use protocols adhere to the German Tierschutzgesetz , the Tierschutz-Versuchstierverordnung and the recommendations of GV-SOLAS .
For direct pathogen detection , DNA was extracted from murine blood samples and the flaB gene of B . persica was detected with a real-time PCR . Data are shown as box plots using a log10-scale of the absolute spirochete numbers per ml blood ( Y-axis ) and plotted against the blood sampling days ( X-axis; Fig 1A–1C ) . When mice were inoculated intradermally with 1 x 106 B . persica organisms ( Fig 1A ) , spirochetes were detectable in their blood starting one day after injection . Median spirochete concentration ranged from 4 . 80 to 6 . 59 ( log10 x organisms/ml ) during the first 12 days . A substantial decline in detectable spirochete numbers was observed from day 12 to 14 . The median spirochete numbers dropped from 4 . 80 to 0 ( log10 x organisms/ml ) . After day 14 , the majority of the mice tested negative for spirochetes in the blood , while during the same period three mice showed reduced numbers of borrelia and only one of them produced a positive signal on the final day of the experiment . The highest spirochete burden observed in a blood sample of an individual mouse was 1 . 9 x 107 B . persica/ml . When less spirochetes were used for intradermal inoculation ( 1 x 104 and 1 x 102 B . persica/mouse ) , spirochetes appeared in the blood circulation of the mice with a delay compared to the experiment performed with 1 x 106 B . persica/mouse . When 1 x 104 B . persica organisms were injected ( Fig 1B ) , the earliest spirochetes were detectable two days after inoculation . Three peaks in spirochete concentration were recorded until day 15 ( median spirochete concentration ranged from 3 . 90 to 6 . 43; log10 x organisms/ml ) and then the spirochete number decreased from day 15 onwards ( 6 . 31 to 0; log10 x organisms/ml ) . After day 16 , the median spirochete concentration varied at a low level ( 1 . 60 to 4 . 21; log10 x organisms/ml ) . The majority of the mice tested negative from day 26 onwards . The highest spirochete load observed in a blood sample of an individual mouse was 2 . 3 x 107 organisms/ml . When 1 x 102 B . persica organisms were injected ( Fig 1C ) , spirochetes were detectable beginning on day 3 . Varying spirochete numbers in blood were observed during the first 16 days of the experiment and from day 17 onwards the median number of spirochetes decreased substantially ( from 5 . 28 to 2 . 01; log10 x organisms/ml ) . Mice showed spirochetemia at a low level until day 20 . Then , the majority of the animals tested negative . No signals for flaB DNA were recorded for mice that were exposed to only four B . persica organisms per mouse and from mice which served as negative controls . In summary , intradermal injection of decreasing numbers of B . persica resulted in delayed appearance of the spirochetes in the blood of infected mice . After an initial fluctuation of the median spirochete concentration at a high level , a sudden decrease in spirochete numbers was observed in each group from day 13 to 18 . The infection rate was 100% when mice were injected intradermally with doses of 1 x 106 ( 20/20 ) , 1 x 104 ( 8/8 ) and 1 x 102 ( 8/8 ) B . persica/mouse . The lowest dose tested ( four B . persica/mouse ) did not result in infection of any mouse ( 0/8 ) . When analyzed at the individual level , most animals showed two to three peaks of spirochetemia and only a few ( 3/36 ) produced one peak . Fig 2A–2C show the absolute spirochete numbers per ml blood of three selected mice . These individual animals , which received 1 x 106 B . persica/mouse , revealed three different relapse patterns . In terms of mouse #2 ( Fig 2A ) , peak spirochetemia intensities declined over time . First , spirochete numbers increased up to 2 . 2 x 106 organisms/ml on day 3 . Subsequently , spirochetemia intensity decreased until day 5 , rose up to 1 . 3 x 106 organisms/ml on day 7 , declined again on day 9 and showed a last small peak of 6 . 0 x 105 organisms/ml on day 11 . After that , the animal was negative until the end of the experiment . The lowest detectable number of spirochetes between the peaks was recorded on day 9 ( 1 . 2 x 105 organisms/ml ) . In mouse #12 ( Fig 2B ) , two peaks ( 5 . 3 x 106 organisms/ml on day 3 and 4 . 7 x 106 organisms/ml on day 7 ) were noted . Between the peaks the detectable spirochete concentration was 1 . 2 x 106 organisms/ml . This mouse was negative after day 13 . Fig 2C depicts the kinetics of spirochetemia in mouse #18: the first high peak with 7 . 1 x 106 organisms/ml on day 2 was followed by a second low peak with 3 . 8 x 106 organisms/ml on day 6 , which again was followed by a last high peak with 7 . 4 x 106 organisms/ml on day 10 . Between the first and the second peak the lowest spirochete count was 1 . 1 x 106 organisms/ml on day 4 , while only 1 . 7 x 105 organisms/ml were observed between the second and the third peak on day 8 . This mouse also remained negative from day 12 onwards . Interestingly , none of the infected mice showed any clinical signs or elevated temperatures during spirochetemia and the following periods when compared to negative control animals . Tissue samples were collected from the skin around the infection area , heart , spleen , urinary bladder , left tarsal joint , and brain at the end of the infection experiment with 1 x 106 borrelia/mouse on days 49/50 . Cultures with liquid medium were started to attempt the cultivation of the borrelia . The organ cultures were investigated for the presence of viable spirochetes under a dark-field microscope once a week over a period of three weeks . One week after initiating the cultures , at least some single borrelia organisms were observed in most brain cultures . After three weeks , 13 out of 20 observed brain samples were found positive with some cultures showing massive numbers of rapidly-moving spirochetes ( Table 1 ) . Three skin cultures were also observed as positive after three weeks . In total , 70% ( 14/20 ) of the infected mice tested positive by culture . By real-time PCR , 90% ( 18/20 ) of the infected mice tested positive for the B . persica flaB gene in tissue samples . In addition to 18 positive brain and three skin samples from 18 mice , one heart and one splenic sample of the same mouse were positive by real-time PCR ( this mouse was also positive by blood real-time PCR performed on its final day of experiment ) . Control animals tested negative in both methods . According to the results , only brain and skin samples were investigated on the final days of the dose finding study . Spirochetes were seen in 87 . 5% ( 7/8; infection dose: 1 x 104 B . persica/mouse ) and 100% ( 8/8; infection dose: 1 x 102 B . persica/mouse ) of brain cultures . All tissue cultures of animals that had received four B . persica/mouse and all negative controls tested negative . Real-time PCR was 100% ( 8/8 ) positive for brain samples from mice infected with doses of 1 x 104 and 1 x 102 B . persica/mouse . None of the skin samples tested positive in both methods . Five mice ( inoculated with 1 x 106 B . persica/mouse ) which were positive according to all other test methods were selected for histopathologic evaluation . A negative mouse served as a control . The paraffin-embedded slices of brains and kidneys as well as the plastic-embedded slices of brains and joints revealed no histopathological changes indicative for inflammatory responses . Two specimens of in plastic-embedded kidneys ( one infected and the uninfected mouse ) contained small scattered interstitial infiltrations of lymphocytes ( mild interstitial non-suppurative focal nephritis ) . The specific antibody response against B . persica was measured with a kinetic ELISA and characterized by western blotting . Plasma samples of mice injected with 1 x 106 B . persica/mouse were collected and pooled according to subgroups from day 1 to 50 according to an alternating sample collection schedule . Plasma samples as well as individual final serum samples of all animals were tested with an ELISA for the detection of murine IgM , IgG and IgA antibodies . Antibody levels developed immediately after spirochete injection and rose to 381 . 6 KELA units until day 21 . Antibody levels plateaued ( 416 . 4 to 479 . 5 KELA units ) until day 50 ( Fig 3A ) . Antibody levels of individual final serum samples are shown in Fig 3B . The highest antibody levels were obtained in animals injected with 1 x 106 B . persica/mouse on days 49/50 of the experiment . Mice exposed to 1 x 104 or 1 x 102 B . persica/mouse showed medium to high levels of specific antibodies , however their antibody levels were lower when compared to the high-dose exposed group on day 28 ( pooled plasma samples ) . Sera of animals receiving only four B . persica/mouse and negative mice showed non-specific antibody responses . Western blots of individual final sera from each infection dose group showed bands between 15 kDa and 100 kDa . The patterns of the infected mice were similar . Nevertheless , the lower antibody levels induced by the smaller infection doses were reflected in the intensities of the immunoblot bands . Negative controls and animals that had received four B . persica/mouse showed only non-specific bands at 25 kDa and 37 kDa ( Fig 3C ) .
The results of this study show that B . persica strain LMU-C01 can be used to establish infection in immunocompetent C3H/HeOuJ mice . The minimal infectious dose was between four and 1 x 102 B . persica organisms by intradermal inoculation in this study . Spirochetes were detected in the blood , brain and skin tissue samples thereby defining the brain and the skin as target organs of B . persica dissemination . The infection model presented in this study can serve as a platform for further ensuing in vivo investigations to gain new insights into the pathogenesis of B . persica . | The spirochete Borrelia persica is a tick-borne bacterium that is transmitted by the vector Ornithodoros tholozani to its vertebrate host in the Middle East , Central Asia and the Indian peninsula . Current migration of vast numbers of individuals from this area increases the likelihood that B . persica infections will be introduced into new geographic regions . After infection and distribution by the bloodstream , relapsing fever episodes occur in humans . Since no reliable in vivo tools have existed so far to study this organism , a murine model was established in this study to characterize the infection kinetics in immunocompetent mice . Aspects of the potential infectivity of the laboratory strain and of potential clinical signs , spirochetemia and antibody response as well as organ tropism and histopathological reactions were studied . With the successful infection model presented here , further studies are conceivable in order to gain advanced insights into the pathogenesis of B . persica infection and to characterize in detail the host immune response mounted against the bacterium . We propose that this model could also be used for the development of new rapid diagnostic approaches to initiate or monitor treatment regimes in order to clear or prevent the infection with B . persica . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"borrelia",
"infection",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"enzyme-linked",
"immunoassays",
"pathology",
"and",
"laboratory",
"medicine",
"immune",
"physiology",
"pathogens",
"immunology",
"microbiology",
"animal",
"models",
"bacterial",
"diseases",
"model",
"organisms",
"antibodies",
"immunologic",
"techniques",
"bacteria",
"bacterial",
"pathogens",
"research",
"and",
"analysis",
"methods",
"immune",
"system",
"proteins",
"infectious",
"diseases",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"immunoassays",
"mouse",
"models",
"borrelia",
"hematology",
"blood",
"plasma",
"biochemistry",
"blood",
"spirochetes",
"anatomy",
"physiology",
"biology",
"and",
"life",
"sciences",
"organisms"
] | 2016 | Borrelia persica Infection in Immunocompetent Mice - A New Tool to Study the Infection Kinetics In Vivo |
Ataxia represents a pathological coordination failure that often involves functional disturbances in cerebellar circuits . Purkinje cells ( PCs ) characterize the only output neurons of the cerebellar cortex and critically participate in regulating motor coordination . Although different genetic mutations are known that cause ataxia , little is known about the underlying cellular mechanisms . Here we show that a mutated axJ gene locus , encoding the ubiquitin-specific protease 14 ( Usp14 ) , negatively influences synaptic receptor turnover . AxJ mouse mutants , characterized by cerebellar ataxia , display both increased GABAA receptor ( GABAAR ) levels at PC surface membranes accompanied by enlarged IPSCs . Accordingly , we identify physical interaction of Usp14 and the GABAAR α1 subunit . Although other currently unknown changes might be involved , our data show that ubiquitin-dependent GABAAR turnover at cerebellar synapses contributes to axJ-mediated behavioural impairment .
A number of heterogeneous hereditary and non-hereditary disorders lead to ataxia characterized by coordination failures [1] , [2] , [3] . The spontaneous axJ mutation affects the locomotory system , causing an ataxic phenotype in mice [4] . The mutated gene encodes the deubiquitinating enzyme ( DUB ) Usp14 [5] , a member of the ubiquitin-specific protease family [6] , [7] . Due to insertion of an intracisternal A-particle into intron 5 , expression levels of full-length Usp14 in brains of axJ mice are reduced to about 5% [5] . Usp14 catalyzes the hydrolysis of isopeptide bonds in ubiquitin-protein conjugates [8] . Upon alternative splicing of exon 4 , two isoforms of Usp14 are generated . The full-length isoform contains an addition of 33 amino acids , required for proteasome binding . Accordingly , binding of Usp14 to the proteasome is thought to be necessary for efficient hydrolyse activity of Usp14 [9] , [10] . AxJ mice display an exclusive downregulation of the full-length isoform , thereby representing a specific knockdown of the proteasome binding form of Usp14 . Although the proteasome is likely to be involved in the neurological dysfunctions [9] , Usp14 is unable to process polyubiquitin chains [11] . Since , its physiological substrate is thought to be mono- or oligoubiquitinated [5] , rather than representing a polyubiquitinated protein destined for degradation at the proteasome , Usp14 may have several functions in ubiquitin-signaling pathways . Ubiquitination is a key process in the regulation of synapse formation and function [6] , [7] , [12] . Following endocytosis , ubiquitinated receptors are sorted for lysosomal degradation , thereby preventing their recycling to the plasma membrane [13] , [14] , [15] . For instance , the surface expression of glycine receptors ( GlyRs ) depends on ubiquitination , suggesting an important role for this process in the regulation of synaptic receptor levels [16] . Moreover , ubiquitination of inhibitory GABAA receptors ( GABAARs ) has recently been shown to be activity-dependent and to regulate synaptic GABAAR accumulation [17] . GABAARs mediate the majority of fast synaptic inhibition in the mammalian brain . In the cerebellum , 75% of all GABAARs contain the α1 subunit [18] , whereas PCs exclusively express α1-containing GABAARs [19] , [20] , [21] . PCs transform excitatory afferent signals to inhibitory efferents that target the neurons of the deep cerebellar nuclei ( DCN ) and vestibular nuclei ( VN ) [1] , [22] , [23] . Their inhibitory influence on DCN and VN neurons is a prerequisite for normal motor coordination , and even minor disturbances of cerebellar inhibition has been shown to cause uncoordinated movements and ataxia [1] . Hence , mouse mutants characterized by Purkinje cell degeneration , such as pcd or leaner mice suffer from ataxia [24] , [25] , [26] , [27] . Here , we show that the downregulation of Usp14 in axJ mice is accompanied by a marked redistribution of intracellular α1-containing GABAARs to PC surface membranes , leading to enlarged IPSC amplitudes . We further demonstrate physical interaction of Usp14 and GABAAR α1 , suggesting that Usp14 directly participates in the regulation of synaptic GABAAR turnover . Consistently , interference with GABAAR-Usp14 binding in a heterologous system mimics the in vivo observations . Our data demonstrate a new concept with the ubiquitin-proteasome system ( UPS ) representing a key player in synaptic neurotransmitter receptor regulation .
Mice carrying the axJ mutation display reduced expression levels of the full-length Usp14 isoform in brain , whereas expression of the short Usp14 isoform remains unaltered ( [5] , Figure 1A ) . Phenotypically , axJ mice demonstrate severe coordination failures and ataxia ( Figure 1B ) [4] , often linked to dysfunctions within cerebellar circuits [28] . Although increased apoptotic cell death is reported in axJ-derived cerebellar granule cell layers [29] , application of nuclear staining revealed that the overall architecture of the cerebellum remains normal ( Figure 1C ) . To visualize PC bodies ( arrows ) and the molecular layer , representing PC dendrites , the PC marker protein Calbindin was immunolabeled . ( Figure 1D , upper panels , green ) . Parallel staining of GABAARs using antibodies specific for the α1-subunit ( Figure S1A-S1D ) that represents the only α-type subunit in PCs , demonstrated strongly increased GABAAR cluster intensities in the molecular layer of axJ compared to wt cerebellum ( Figure 1D , upper panels , red; lower panels , white ) . This phenomenon appeared to be mosaic and is in agreement with previous reports of variable expression levels of Usp14LF throughout different tissues [5] . Notably , these effects were specific and due to decreased Usp14 levels , since neuron-specific transgenic expression of Usp14 on the background of axJ mice [9] genetically reversed this effect , thereby leading to similar GABAAR α1 signal intensities as detected in wt PCs ( Figure 1D , axJ x tg , right ) . Together , these data indicate that Usp14 regulates either the gene expression or the subcellular distribution of α1-containing GABAARs in PCs . To investigate the underlying mechanism of increased GABAAR clusters , PCs of wt and axJ mice were analyzed at the subcellular level . Immunostaining of GABAAR α1 using either fluorophore- ( Figure 2A ) or biotin-labeled ( Figure 2B ) secondary antibodies revealed a marked increase in GABAAR α1 clusters at the surface of cell bodies and proximal dendrites ( Figure 2A and 2B , arrows ) . At the ultrastructural level , electron microscopy confirmed that large areas of the axJ PC surface , including extrasynaptic sites , were covered by α1-containing GABAARs ( Figure 2C ) . Notably , the cytoplasm of PCs did not show increased signal intensities between the genotypes ( Figure 2A–2C ) . In addition , western blot analysis of cerebellar protein extracts from wt and axJ mice ( Figure 2D ) as well as mRNA levels upon in situ hybridization ( Figure 2E ) demonstrated equal signals of GABAAR α1 proteins and mRNAs in both genotypes , indicating that the total gene expression of GABAAR α1 is not increased . We therefore conclude that a major loss of Usp14 expression leads to a surface redistribution of intracellular α1-containing GABAARs rather than to a significant increase in GABAAR α1 expression levels . Consistent with the immunohistochemical data , analysis of IPSCs ( n = 10 , 000 events ) indeed revealed that 67% of axJ PCs displayed a significant increase in GABAergic current amplitudes ( Figure 3A–3D , and Figure S2A and Figure S2B ) . In parallel and as expected for a postsynaptic receptor phenomenon , the kinetic parameters , such as rise-time ( 10–90% ) and decay-time ( τ ) remained unaltered under these conditions , indicating that both genotypes display no major changes in neurotransmitter uptake or release mechanisms ( Figure S2C ) . However , the maximal amplitudes in axJ animals ( >150 pA ) displayed significantly ( p = 0 . 005 ) higher decay time constants ( τ = 12 . 1±3 . 1 ms ) , as compared to the decay time constants ( τ = 10 . 3±2 . 9 ms ) of maximal IPSC amplitudes in wt animals ( 80–150 pA; Figure S2C ) . Such differences are consistent with increased perisynaptic receptor numbers and support the immunochemical and EM observations . We therefore conclude that increased GABAAR levels at PC plasma membranes induce increased cerebellar inhibition , that leads to reduced inhibitory output levels of PCs . Notably , also PC degeneration ( pcd ) mutant mice display a severe decrease of PC inhibitory output and develop ataxia [1] , [27] . In addition , GAT1 deficient mice , represented through prolonged GABA actions , due to disturbed GABA reuptake , suffer from ataxia [30] . Hence , altered inhibitory input to PCs leads to similar behavioral consequences compared to the loss of cerebellar inhibitory output upon Purkinje cell degeneration . However , if the observations in the present study contributed to the ataxia phenotype of axJ mice , one should identify a molecular link between GABAAR turnover and Usp14-mediated pathways . To determine , whether Usp14 and GABAAR α1 might physically interact , we applied the lexA-based MATCHMAKER yeast two-hybrid system using Usp14LF ( prey ) and the large intracellular loop of GABAAR α1 ( aa 334–420 , bait ) . These experiments indeed revealed that Usp14 represents a direct GABAAR α1 binding partner ( Figure 4A ) . Fine mapping , using systematic GABAAR α1 deletion mutants , identified the Usp14 binding region within the first 13 amino acids of the α1 loop sequence ( aa 334–346 ) ( Figure 4A ) . Vice versa , the GABAAR α1 binding site within the Usp14 protein was localized at its C-terminal domain ( Figure 4B ) . In order to biochemically verify this interaction , we then applied a pull-down experiment using the GST-tagged GABAAR α1 loop ( aa 334–420 ) . Endogenous Usp14 protein derived from mouse brain lysates specifically bound to the immobilized GST-tagged GABAAR α1 loop , but not to GST alone ( Figure 4C ) , indicating in vitro binding of the protease and the receptor polypeptide . Differential centrifugation of brain extracts revealed that both endogenous proteins cofractionate at P2 plasma-membrane ( 10 , 000×g ) , P3 vesicular ( 100 , 000×g ) , and P4 protein complex ( 400 , 000×g ) fractions . However , while GABAARs are enriched at the plasma membrane ( P2 ) , Usp14 binds to the proteasome and is consequently enriched in fraction P4 ( Figure 4D ) . This marginal overlap is consistent with a transient enzyme-substrate complex , however turned out not to be sufficient to obtain coimmunoprecipitation under standard conditions . Nevertheless , a GFP-tagged Usp14 mutant ( GFP-Usp14 ( H434A-D450A ) ) , harboring two point mutations within its functional catalytic domain [31] , stabilized the complex , and enabled coprecipitation of both full-length binding partners derived from HEK293 cells ( Figure 4E ) . Together , these data demonstrate physical interaction of GABAAR α1 and the ubiquitin-specific protease Usp14 , and suggest that the observed GABAAR redistribution in ataxia mice ( Figures 1 and 2 ) is directly caused by the loss of Usp14 , thereby indicating that GABAAR turnover is ubiquitin-dependent . We next asked whether both proteins colocalize at the subcellular level . While GABAAR α1 subunits have been extensively characterized in both tissues and cells [32] , [33] , immunohistochemical analysis of Usp14 displayed a wide distribution in all layers of the cerebellum ( Figure 5A ) and was detected at synaptic vesicle protein 2 ( SV2 ) -positive synaptic sites in both cultured hippocampal and cerebellar neurons ( Figure 5B , and Figure S1E , turquoise ) . For analysis at ultrastructural resolutions , we performed immunoelectron microscopy using biotin- ( Figure 5C , upper panel ) or gold-labeled ( Figure 5C , middle and lower panels ) secondary antibodies . In this assays Usp14 was detected in close proximity to and directly at postsynaptic sites ( Figure 5C , upper panel , arrows ) , at both the pre- and post-synapse , as well as directly at synaptic plasma membranes ( Figure 5C , middle and lower panels , arrowheads ) . In addition and in consistence with the in vitro binding data , coimmunostaining with antibodies specific for Usp14 and the GABAAR α1 subunit revealed partial colocalization in both cultured hippocampal and cerebellar neurons ( Figure 5D , and Figure S1F , yellow , white arrows ) . At ultrastructural levels , this could be confirmed using gold-labeled secondary antibodies of different particle sizes . In accordance to the literature [33] , GABAAR α1 ( black arrows ) was localized opposed to unlabeled ( Figure 5E , left , arrows ) or synaptophysin-positive presynaptic boutons ( Figure 5E , middle , arrows ) , while colabeling of Usp14 and GABAAR α1 was rather detected at submembrane tubular organelles ( Figure 5E , right , white arrow ) , described in both dendritic shafts and spines [34] . In addition to the smooth endoplasmic reticulum ( SER ) , tubular compartments are generated through merge of internalized vesicles and multivesicular body ( MVB ) -tubule complexes and serve as intracellular stores of material destined for recycling or degradation [34] , [35] , [36] , [37] . Given the fact that organelles that mediate neurotransmitter receptor sorting are localized subsynaptically [34] , [38] with ubiquitin serving as a signal for internalization [39] , [40] , both the observed in vitro binding ( Figure 4 ) and colocalization data ( Figure 5 ) suggest that Usp14 represents a direct regulator of GABAAR turnover . To investigate whether GABAAR α1 might be a putative substrate for Usp14 , we examined whether this subunit could be ubiquitinated in cells . Thus , HEK293T cells were transfected with GFP-tagged GABAAR α1 , GABAAR β3 , HA-tagged ubiquitin and either Usp14 wildtype ( wt ) or a catalytic mutant of Usp14 , respectively ( Figure S4 ) . Extracts of untransfected HEK293T cells served as controls . Upon immunoprecipitation , using anti-GFP antibodies , GABAAR α1-GFP was precipitated from extracts containing GABAAR α1-GFP ( Figure S4A , lower panel ) . Upon the use of HA-antibodies ubiquitinated forms of GABAAR α1 could be detected in extracts from transfected but not untransfected HEK293T cells ( Figure S4A , upper panel ) . The detection of ubiquitinated GABAAR α1 is in line with a recent publication that reported ubiquitinated GABAAR β subunits [17] , suggesting that GABAARs in general are subject to ubiquitin conjugation . In particular the abundance of ubiquitinated GABAAR α1 forms between 75 and 100 kDa ( Figure S4A , upper panel , asterisk ) is slightly increased in the presence of the Usp14 catalytic mutant , represented by a more intensive blurred signal ( see magnified image in Figure S4B ) . This observation suggests a stabilization of mono-/oligoubiquitinated GABAAR α1 polypeptides upon binding of a functionally inactive form of Usp14 . Thus , Usp14 might represent a critical DUB to GABAAR α1 . A balanced control of GABAAR ubiquitination and deubiquitination might therefore be an important determinant in regulating GABAAR surface expression in neurons . If the above interpretations were true , a minimal heterologous system should verify that Usp14 directly affects GABAAR turnover . In addition to the loss of Usp14 in mice , we aimed to proof , whether heterologous overexpression of an isolated Usp14 binding site ( compare with Figure 4A ) of GABAAR α1 would mimic the receptor surface distribution phenotype upon competitive interference with GABAAR α1–Usp14 binding . Thus , HEK293 cells were transfected with constructs encoding GABAAR α1-GFP , GABAAR β3 and the monomeric red fluorescent protein ( mRFP ) -tagged Usp14 binding site of GABAAR α1 ( mRFP-GABAAR α1 ( 334–346 ) ) or with mRFP , respectively . Biotinylation of surface proteins , followed by immunoprecipitation , indeed revealed a 2 . 5-fold increase of GABAAR α1-GFP surface membrane levels in the presence of the competing peptide ( Figure 6A and 6B ) , thereby leading to the same functional consequence , as observed in axJ mice . To verify both the expression and catalytical activity of Usp14 in HEK293 cells , we performed western blot analysis of protein extracts from kidney and cultured HEK293 cells , using a HA-tagged ubiquitin vinyl methyl ester ( HAub-VME ) , active site probe [41] , [42] . Western blot analysis using HA-specific antibodies confirmed that HEK293 cells express catalytically active Usp14 ( Figure S3A ) . Hence , we conclude that the disruption of GABAAR α1-Usp14 binding , and consequently the gene expression knockdown of Usp14 in axJ mice , is directly causal for increased GABAAR surface membrane expression , known to result in enlarged IPSC amplitudes in vivo . Since Usp14 represents a protease , its enzymatic activity should also be critical in this respect . To test this , we transfected HEK293 cells with constructs encoding GABAAR α1-GFP , GABAAR β3 and a catalytically inactive Usp14 mutant or with mRFP , respectively . Overexpression of the loss-of-function mutant , notably resulted in 3-fold enrichment of GABAAR α1-GFP surface expression ( Figure 6C and 6D ) , as compared to control experiments , indicating that a balanced turnover of α1-containing GABAARs indeed requires a catalytically intact Usp14 enzyme . In summary , these and other data in this study suggest that the function of Usp14 is directly involved in GABAAR turnover . Regulation of synaptic strength requires the precise control of neurotransmitter receptor numbers at synaptic sites . Our in vivo and in vitro observations in this study indicate the novel concept that DUB-dependent pathways regulate neurotransmitter receptor density and might participate in synaptic plasticity mechanisms . Since axJ mice represent mutants that are exclusively deficient for the proteasome-associated form of Usp14 , with deubiquitinating enzymes playing an important role in the UPS , our data suggest a role of the proteasome in GABAAR turnover . It has been shown , that epidermal growth factor receptors ( EGFRs ) , once activated , undergo ubiquitination and internalization from the plasma membrane [43] . Prior to their sorting into multivesicular bodies ( MVBs ) , EGFRs require deubiquitination , a process that depends on proteasomal activity , although EGFRs , such as most transmembrane proteins , undergo degradation at lysosomes . Recent data in yeast further confirm a proteasomal contribution in similar processes , by showing that the proteasome-associated deubiquitinating enzyme Doa4 removes ubiquitin from cargo proteins prior to their entry into internal vesicles of MVBs [14] , [44] , [45] . Usp14 , as it binds to the proteasome and mediates deubiquitination , is therefore a candidate factor to serve in similar pathways in neurons ( Figure 6E ) . Consequently , Usp14 might represent the responsible DUB to trigger GABAAR transport from early endosomes into MVBs/lysosomes , although it might already interfere with GABAARs at the cell surface or right after internalization ( Figure 6E ) . Interruption of this function , either by loss of Usp14 ( axJ mice ) , or by interference with Usp14-GABAARs α1 binding ( HEK293 cells ) , might either induce ( i ) backpropagation of disturbed endocytic pathways , leading to maintenance of the receptor at the cell surface or ( ii ) increased recycling of receptors back to the cell surface . Although an exact molecular mechanism remains to be elucidated , Usp14 represents a novel candidate to participate in the regulation of GABAAR turnover and synaptic plasticity at GABAergic synapses . Since axJ animals show further neuronal abnormalities , such as impaired synaptic transmission at neuromuscular junctions , their neurological phenotype might be due to a combination of deficits . However , we conclude that impaired GABAAR turnover in PCs due to the loss of Usp14 significantly contributes to the ataxic phenotype in axJ mice . This view is supported by previous studies , which report that mouse mutants with altered GABAAR densities or reduced GABAergic terminals in the cerebellum , also develop severe motor impairments [1] , [25] , [27] , [46] . For instance the pcd mouse mutant , characterized by a complete loss of PCs , shows ataxia . PC degeneration in pcd mice results in a loss of inhibitory PC output and consequently a reduced inhibitory input to the vestibular nuclei , representing one of the direct PC target regions . In addition , altered Purkinje cell input leads to ataxia . Hence , mice deficient for the GABA transporter GAT1 develop severe ataxia due to disturbed GABA re-uptake . Consequently , this functional deficit leads to an increased GABAA receptor-mediated tonic conductance and prolonged IPSCs in both cerebellar granule and Purkinje cells [30] . It is therefore a possible scenario that motor impairments are closely linked to the level of inhibition within cerebellar circuits . The observed increase in GABAAR surface expression and IPSC amplitudes in axJ mice , as reported in this study , negatively affect PC functions and are likely to contribute to the ataxic symptoms . In parallel other yet unknown changes might be involved , too . Hence , in addition to GABAergic transmission , ubiquitin-mediated pathways might be putative targets for therapeutic treatment against certain forms of cerebellar ataxia .
The entire cDNA sequence of GABAAR α1 was subcloned as an EcoRI/SalI fragment into the pEGFP-N vector ( BD Biosciences ) . The cDNA of Usp14 was subcloned as a BamHI fragment into the pFLAG-CMV-2 vector ( Sigma ) . HA-tagged Ubiquitin was subcloned as an EcoRI/XhoI fragment into the pcDNA3f vector ( Invitrogen ) . Single mutations ( Usp14: C114A; GABAAR α1 loop: introduction of stop codons to generate deletion mutants and the competitive peptide , respectively ) or group mutations ( Usp14: H434A-D450A ) were introduced using the site-directed mutagenesis kit ( Stratagene ) . Immunofluorescence: rabbit anti-GABAAR α1 ( 1∶250 Upstate ) ; guinea pig anti-GABAAR α1 ( 1∶6000 ) ; rabbit anti-Usp14 [10] rabbit anti-Usp14 ( 138-R/PB+2 , SM Wilson-lab . ) ; mouse anti-Calbindin ( 1∶100 , Sigma ) ; mouse anti-synaptic vesicle ( SV2 , 1∶100 , Hybridoma bank , University of Iowa ) ; Secondary antibodies: CY3- , CY2- or CY5-conjugated donkey-anti rat , mouse , guinea pig or rabbit ( all 1∶500 , Dianova ) . Immunoprecipitation/Western blot analysis: rabbit anti-GABAAR α1 ( 1∶500 Upstate , 1∶1000 AbD serotec ) ; guinea pig anti-GABAAR α1 ( 1∶6000 JM Fritschy-lab . ) ; mouse-anti-Usp14 ( IA4; 1∶1000 , SM Wilson-lab . ) mouse anti-pan Cadherin ( 1∶100 , Abcam ) ; mouse anti-N-Cadherin ( 1∶4000 , Cell Signaling Technology ) ; rabbit anti-actin ( 1∶2000 , Sigma ) ; mouse anti-HA ( 1∶1000 , Sigma; 1∶1000 , Santa Cruz Biotechnology ) ; anti-GFP ( 1∶1000 , Roche ) , mouse anti-rpt4 ( 1∶1000 , Biomol ) Secondary antibodies: HRP-conjugated goat-anti rabbit , guinea pig and mouse ( all 1∶10 . 000 , Dianova ) ; HRP-conjugated protein A ( 1∶1000 , KPL ) ; biotinylated secondary antibodies ( 1∶1000 , Vector laboratories ) , gold-labeled secondary antibodies . Preembedding immunocytochemistry: mice were anaesthetized and perfused with 4% PFA with 0 . 1% glutaraldehyde in PBS . Sagittal vibratome sections of the cerebellum were cut ( 60 µm ) . After washing in PBS , sections were treated with 0 . 3% H2O2 and 1% NaBH4 in PBS for 30 min . After rinsing in PBS , sections were incubated with 10% horse serum ( HS ) containing 0 . 2% BSA for 15 min and subsequently incubated over night with primary antibodies in PBS , containing 1% PS and 0 . 2% BSA ( carrier ) . Sections were washed in PBS , incubated with biotinylated secondary antibody and diluted in carrier for 90 min . After rinsing , sections were incubated with ABC ( Vector Labs ) and diluted to a 1∶100 concentration in PBS for 90 min . Afterwards they were washed in PBS and further incubated in diaminobenzidine ( DAB ) -H202 solution ( Sigma ) for 10 min . Sections were then either mounted on glass coverslips ( light microscopy ) or postfixed with 1% OsO4 , dehydrated in an ascending series of ethanol and embedded in Epon ( Roth ) . Ultrathin sections were examined with a Zeiss EM 902 . Immunocytochemistry of ultrathin frozen sections: mice were perfused and cerebellar sections of the cerebellum were cut ( 100–200 µm ) , as described above . Small blocks of cerebellar tissue containing all layers were immersed in 12% gelatin in PBS at 37°C for 15–30 min . Blocks were transferred into vials containing 2 . 3 M sucrose in PBS and incubated over night . Thereafter they were frozen on specimen holders in liquid nitrogen . Ultrathin sections were prepared at a Reichard Ultracut microtome , equipped with a cryochamber and placed on copper grids ( Sciences Services ) . Single and double immunogold labeling was performed according to Slot and Geuze using secondary 10 nm large protein A gold to label rabbit primary antibodies and 6 nm large gold ( Dianova ) to label guinea pig primary antibodies [49] . Mice were anaesthetized and the cerebellar vermis prepared in ice-cold carboxygenated ACSF ( NaCl 135 mM; KCl , 5 mM; CaCl2 2 mM; MgCl2 1 mM; glucose 10 mM; Na2HCO3 , 30 mM; NaHPO4 , 1 . 5 mM; pH 7 . 4 ( bubbled with carbogen ) ) . The tissue was cut into 200 µm sagittal sections ( Microm HM 650V; Histoacryl glue , Braun ) , that were transferred to carboxygenated ACSF at 35°C for 20–30 min before being kept at RT ( 22–24°C ) , until further use . Slices were placed in a recording chamber ( RC26GLP , Warner Instr . ) under an upright microscope ( BX51WI , Olympus ) . Individual PCs were visually identified and recorded with borosilicate capillaries of approximately 5 MO resistance ( Hilgenberg ) using the whole-cell patch-clamp configuration . Spontaneous synaptic events were recorded under equimolare Cl− concentrations at −60 mV and the GABAergic input isolated using the AMPA-type glutamate antagonist CNQX; the remaining IPSCs could be blocked by 20 µM bicuculline ( Figure S2A ) . IPSC were recorded at 10 kHz for 60 s every 10 min . over approximately 1 h using the Patchmaster 2 . 05 software ( HEKA ) . Data were analysed with the MiniAnalysis 6 . 02 program ( Synaptosoft; converted with the supplied ABF Utility ) using identical parameters for evaluating all IPSCs . Intracellular electrode: CsCl , 125 mM; MgCl2 , 2 mM; EGTA 0 . 1 mM; TEA 5 mM; Na2-ATP , 4 mM; Na-GTP , 0 . 5 mM; HEPES , 10 mM; pH 7 . 3 ( CsOH ) . HEK293 cells were harvested in 1% Triton X-100 , 48 h after transfection . Antibodies were coupled to 30 µl of protein G beads ( Dynal Biotech ) in IP washing buffer ( 50 mM TrisHCl , 150 mM NaCl , 5 mM MgCl2 , PH 7 . 1 ) . Cell extracts were incubated with the beads over night , then washed and boiled in SDS sample buffer . For GST-pulldown experiments , HEK293 cells were harvested 48 h after transfection in 1 ml 1% Triton X-100 . E . coli BL21 lysates were obtained by sonification and centrifugation at 10 , 000×g for 30 min . Bacterial lysates were coupled to glutathione-sepharose beads ( Amersham ) for 3 h . HEK293 cell lysates were applied to the beads for 10–12 h . Beads were washed and boiled prior to Western blot analysis . Proteins separated by SDS PAGE were transferred to PVDF or nitrocellulose membranes and unspecific binding sites were blocked using PBS containing 0 . 1% Tween and 5% skim milk powder . Primary and secondary antibody incubation was performed in blocking solution . HEK293T cells were transfected using GeneJuice Transfection Reagent ( 10 µg DNA/10 cm dish ) . 48–72 h after transfection , cells were lysed in lysis buffer ( 50 mM HEPES , 150 mM NaCl , 10% glycerol , 1 mM EGTA , 1 mM EDTA , 25 mM NaF , 10 µM ZnCl2 pH 7 . 5 ) supplemented with 10 mM NEM to inhibit deubiquitinating enzymes as well as protease and phosphatase inhibitors ( 10 µg/ml Aprotinin; 2 µg/ml Leupeptin; 1 mM PMSF; 1 mM Na-orthovanadate ) . Cell lysates were preincubated with 20–25 µl Protein A/G PLUS Agarose ( Santa Cruz Biotechnology ) to remove unspecifically bound proteins . Immunoprecipitation using anti-GFP antibodies ( 1 . 0–1 . 2 µg; Roche ) was performed overnight at 4°C . Proteins bound to GFP-Antibodies were precipitated by adding 25 µl Protein A/G PLUS Agarose ( Santa Cruz Biotechnology ) followed by incubation for 45 min at 4°C . Precipitates were analysed by western blot analysis as described above considering the following differences . Unspecific binding sites were blocked using TBS ( 150 mM NaCl , 50 mM Tris , 0 . 1% Na-Azide , 0 . 5% v/v phenol red ) containing 5% BSA . Primary antibodies were diluted in TBS/5% BSA . 48 h after transfection , HEK293 cells were incubated ( 20 min; 4°C ) with HEPES containing 1 mM biotinamidohexanoic acid 3-sulfo-N-hydroxysuccinimid-ester sodium salt ( Sigma ) . Remaining biotin reagent was quenched by adding 100 mM glycine ( twice for 20 min at 4°C ) . Cells were washed with ice cold PBS and lysed in PBS containing 1% Triton X-100 and protease inhibitor cocktail ( MiniComplete 1 tablet/10 ml Roche ) . After a 30 min incubation step on ice , followed by a brief centrifugation-step at 1 , 000×g ( 5 min , 4°C ) , 30 µl of the supernatants were loaded on a gel to evaluate the amount of GABAAR α1-GFP . After quantification adjusted volumes of supernatants were added to 30 µl of prewashed magnetic Streptavidin MyOne beads ( Dynal ) to achieve equal amounts of GABAAR α1-GFP used for precipitation . Beads were incubated at 4°C for 3 h on a rotation wheel , washed 3 times , collected and boiled in SDS sample buffer . For protein-protein interaction analysis , the Matchmaker LexA yeast Two-Hybrid system ( Clontech , Heidelberg , Germany ) was used . Interactions of bait ( pGilda ) and prey ( pJG4-5 ) fusion proteins were examined by activation of a LEU2 and a lacZ reporter gene [47] . For detection of GABAAR α1 mRNAs , antisense oligonucleotides were synthesized encoding the large intracellular loop region between transmembrane domains M3 and M4 ( aa 342-356 ) [51] . In situ hybridization was performed as previously described [51] , [52] . | Cerebellar ataxia describes a combination of motor symptoms and uncoordinated movements that originates from various hereditary and non-hereditary diseases . Although functional disturbances of cerebellar inhibitory output signals are thought to cause ataxia , the underlying molecular mechanisms are barely understood and medical treatment therefore remains difficult . We analysed a behavioural abnormality up to the molecular level in a mouse mutant ( axJ ) representing a model for ataxia . The axJ mutation reduces the expression level of a ubiquitin protease ( Usp14 ) leading to an abnormal turnover of neurotransmitter receptors . Despite other yet unknown changes in axJ mutants , our data show that intracellular protein turnover contributes to a motor behavioural syndrome . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Methods"
] | [
"cell",
"biology/membranes",
"and",
"sorting",
"neuroscience",
"genetics",
"and",
"genomics/disease",
"models",
"neurological",
"disorders/movement",
"disorders"
] | 2009 | The Ataxia (axJ) Mutation Causes Abnormal GABAA Receptor Turnover in Mice |
Since the late 1980s , wild salmon catch and abundance have declined dramatically in the North Atlantic and in much of the northeastern Pacific south of Alaska . In these areas , there has been a concomitant increase in the production of farmed salmon . Previous studies have shown negative impacts on wild salmonids , but these results have been difficult to translate into predictions of change in wild population survival and abundance . We compared marine survival of salmonids in areas with salmon farming to adjacent areas without farms in Scotland , Ireland , Atlantic Canada , and Pacific Canada to estimate changes in marine survival concurrent with the growth of salmon aquaculture . Through a meta-analysis of existing data , we show a reduction in survival or abundance of Atlantic salmon; sea trout; and pink , chum , and coho salmon in association with increased production of farmed salmon . In many cases , these reductions in survival or abundance are greater than 50% . Meta-analytic estimates of the mean effect are significant and negative , suggesting that salmon farming has reduced survival of wild salmon and trout in many populations and countries .
Since the late 1970s , salmon aquaculture has grown into a global industry , producing over 1 million tonnes of salmon per year [1] . The majority of this biomass is held in open net pens in coastal areas through which wild salmon migrate on their way to and from the ocean . A number of studies have predicted or evaluated the impacts of salmon farming on wild salmon through a single mechanism , in a given area . It is clear that some salmonids are infected and killed by sea lice originating from salmon farms [2–5] , that other diseases have been spread to wild populations from salmonid farming activities [6 , 7] , and there is evidence that salmon parr are at lower density in areas of Scotland where there is salmon aquaculture [8] . In addition , farmed salmon escape in all areas where salmon aquaculture is practiced , and although their breeding success may be low on average , competition for mates and hybridization with wild salmon are likely to reduce survival of wild populations [9 , 10] . It is well established that wild salmonids can be negatively affected by salmon farming [11] , however , the importance of these interactions at the population level has rarely been determined [2] . To determine population level impacts , we examined temporal trends in the abundance and survival of wild salmonids ( Figure 1 and Figure S1 ) . Our study contrasted trends in wild populations exposed to potential aquaculture impacts with those of populations not exposed . Populations in which juvenile salmonids pass by salmon farms during their migration were considered to be exposed to impacts of salmon farming . Exposed populations were carefully paired with control populations in the same region whose migrations did not lead past farms , but which otherwise experienced similar climate and anthropogenic disturbances . Use of such paired comparisons allowed us to control for confounding factors such as climate to detect population level impacts . Using the Ricker stock recruit model [12] , we performed 11 comparisons , involving many stocks from both sides of the Atlantic and from British Columbia in the Pacific ( Table 1 , Data section of Materials and Methods ) .
All estimates of the effect of aquaculture on survival or returns were negative . Both random effects estimates of the mean effect were negative and highly significant ( Figure 2 ) , indicating a very large reduction in survival and returns in populations exposed to aquaculture . Under the dynamics of Equation 1 ( see Materials and Methods ) , percent change in survival or returns is represented by where γ is the coefficient of aquaculture production ( P ) for region k . For example , the estimated change in survival per tonne of salmon farming ( γk ) for Bay d'Espoir in Newfoundland was estimated to be 0 . 026 ( Figure 2 ) . In 2003 , the farmed salmon harvest from this area was 1 , 450 tonnes ( t ) , so the estimated decrease in survival is ( 95% CI: 44%–80% ) , relative to what it would be in the absence of farms . Survival and total returns of many stocks were found to be reduced by more than 50% ( Figure 2 ) , for each generation . If all exposed populations were passing by farms with a total annual harvest of 15 , 000 t , the mean estimated total reduction in survival would be 73% ( 95% CI: 29%–90% ) ( Figure 2 ) . Many regions now have farmed salmon production in excess of 20 , 000 t/y . Generally , Atlantic salmon populations were depressed more than Pacific salmon populations , particularly Atlantic salmon in Atlantic Canada . Irish sea trout were also estimated to have been very strongly reduced by impacts of salmon farming , whereas estimated impacts on Atlantic salmon in Scotland depended on the data used . In British Columbia ( Pacific Canada ) , only pink salmon showed significant declines correlated with salmon aquaculture . Results are reported for a model including autocorrelated errors and with λ set at 0 . 5 , rather than 1 or 2 , because this minimized the Akaike information criteria ( AIC ) for most regions [13] . The parameter λ allows for the impacts of salmon farming to change nonlinearly with the aquaculture production . A λ of 0 . 5 indicates that relatively small amounts of aquaculture will depress wild populations , but the effect does not increase proportionally to aquaculture production . See Tables S1 and S2 for results of alternative models . For the New Brunswick comparison , the outer Bay of Fundy rivers are located much closer to salmon farms than the other exposed rivers . If only these outer Bay of Fundy rivers are considered exposed to salmon farming , and other Bay of Fundy rivers ( inner Bay of Fundy and Saint John River ) are included among the controls , the overall estimates ( i . e . , meta-analytic means ) are still significant and negative in both versions of the analysis .
We have estimated a significant increase in mortality of wild salmonids exposed to salmon farming across many regions . However , estimates for individual regions are dependent on assumptions detailed in the Materials and Methods section , and the estimates often have large confidence intervals . Given that the data analysed are affected by considerable noise—including changes in fishing and environmental factors—the important result of this study is that we are nonetheless able to detect a large , statistically significant effect correlated with trends in farmed salmon production . The significant increase in mortality related to salmon farming that we have estimated in almost all cases is in addition to mortality that is also acting on the control populations . In most cases , control populations were also experiencing decreases in marine ( and sometimes freshwater ) survival , for reasons that are only partially understood . At the same time , fishing mortality has been reduced or eliminated in many areas , which may have partially masked high mortalities associated with aquaculture . A key assumption in this study is that exposed and control areas do not differ in a systematic way across regions . We have identified three possible ways that exposed and control sites could differ systematically: first , salmon farms could be established only in areas where wild stocks have already collapsed; second , salmon farms could be established in areas where habitat is more disturbed by human activities; or , third , climate factors could differ between the exposed areas and the controls in a systematic way . Declines in control and exposed salmonid populations preceded the growth of the salmon aquaculture industry in some regions , but inspection of the data used do not indicate that salmon populations in the majority of our regions had declined dramatically in the exposed areas only , before the start of salmon farming ( averaged returns data are shown in Figure 1 ) . In regions such as Scotland , where declines precede the start of salmon farming , the strong aquaculture effect estimated reflects a faster decline in exposed populations concurrent with the growth of salmon farming . Areas that we consider exposed do not seem to be more developed than control areas in general . In the Atlantic , most areas have been highly altered by human activities for hundreds of years , but there is no obvious difference between the control and exposed groups in this regard . In British Columbia , all areas considered are very remote , and the main type of anthropogenic disturbance in rivers would be forestry . Comprehensive forestry records at the watershed scale are not easily available , but logging in British Columbia's Central Coast is extensive , both historically and recently [14] . It should be noted that the comparisons in British Columbia include large numbers of rivers ( > 80 rivers in each case ) , so differences in anthropogenic effects would have to hold over many watersheds to explain the effects we estimate . Finally , it is also very unlikely that our results are due to a climate driven trend in which more southerly populations show stronger declines than populations to the north . Although our exposed populations are to the south of control populations in three of five regions , differences in latitude are small . In New Brunswick , the control populations are to the north of the exposed populations , but by less than 200 km , and the headwaters of some of the exposed populations are adjacent to those of the controls . In Newfoundland , the difference in latitude between exposed and control populations is similarly small . In British Columbia , the control populations are also to the north , but by less than 300 km . Also , Mueter et al . [15] found that pink and coho salmon from all of the British Columbia populations we have examined respond similarly to large-scale climate trends . Thus , the pattern we found in this study does not seem attributable to a systemic difference between the control and exposed areas . We estimated higher impacts on populations in the Atlantic than those in British Columbia , possibly because Atlantic salmon populations are conspecific with farmed salmon , and therefore susceptible to genetic effects from interbreeding with escaped farm salmon , in addition to disease or other impacts . Estimated impacts in British Columbia may also be lower because we aggregated over large numbers of populations for pink , chum , and coho salmon , because estimates of fishing mortality were only available at a very coarse scale . The individual populations may vary in their exposure to salmon farms . The large apparent impact of Atlantic salmon farming on Irish sea trout , in contrast , can not be explained by interbreeding . In the mid-western region of Ireland ( the exposed region ) , the total rod catch decreased from almost 19 , 000 sea trout in 1985 to 461 in 1990 [16] . In the few rivers where data were available , catch declines could not be explained by reduced effort [16] . Welsh sea trout catches ( the controls ) have remained relatively constant during the same time period , whereas fishing effort has decreased considerably [17] . Sea trout ( anadromous brown trout ) might be expected to experience higher mortalities , because they spend lengthy periods in coastal areas near salmon farms , relative to Atlantic salmon , thus being exposed to disease or parasites for a longer time [18] . The time period over which we are estimating impacts of aquaculture includes the establishment of the industry in each region . Improvements in management as industries mature may explain our finding that impacts of salmon farming on wild salmon do not increase linearly with the tonnage of farmed salmon . Better management should decrease the impact of salmon farming on a per tonne basis , although such improvements may not be able to keep pace with the growth of the salmon farming industry . The estimated reduction in survival of wild salmonids is large , and would be expected to increase if aquaculture production increases .
We analysed data for five species of wild salmonid in five regions: Ireland and Wales , Scotland , Newfoundland ( Canada ) , New Brunswick ( Canada ) , and British Columbia ( Canada ) . There are three further regions with both wild salmonids and salmon aquaculture for which we could not carry out analyses: Norway , the west coast of Vancouver Island ( Canada ) , and Maine ( United States ) . We were unable to carry out analyses for Norway for three reasons . First , salmon farming in Norway is so widespread [21] that it was difficult to establish controls . Second , the adult population in many rivers has been found to contain over 50% aquaculture escapees [22] , making trends in returns to rivers difficult to interpret . Third , there are confounding effects from acidification and disease [23 , 24] . For the west coast of Vancouver Island , it was not possible to obtain aquaculture production data by region over time , and Maine was not included because of a lack of nearby wild populations to serve as controls . Most populations that we considered to be exposed breed in rivers that discharge into bays or channels containing at least one salmon farm . Others breed in rivers flowing into bays without salmon farms very close to areas containing many farms . Salmon from control rivers are very unlikely to pass by salmon farms early in their life cycle , due to the direction of their migration . However , some controls may be relative , in the sense that salmon may pass by farms from a considerable distance , later during their migrations . This would tend to be conservative with respect to our study , since we would then have to detect local effects that are additional to any impacts from distant farms . Data from scientific surveys , e . g . , counting fences , were used if possible; for Scottish salmon and Irish and Welsh sea trout , only catch data were available , so results are given for only the impacts on returns ( not survival ) . We compared rod catches of sea trout in Ireland's Western Region to rod plus in-river fixed engine catches in Wales , from 1985 to 2001 ( there are no fixed engine fisheries directed at sea trout in Ireland ) . Salmon farming is concentrated in the Western Region ( Connemara area ) of Ireland , but does occur in other parts of the country [25] . Based on farm locations [25] , it was estimated that all rivers considered exposed are located less than 50 km from a salmon farm , but most will enter the ocean less than 30 km from a salmon farm . There is no salmon farming in Wales . There were 16 rivers in Western Ireland considered exposed: Athry , Bhinch ( Lower ) , Bhinch ( Middle ) , Bhinch ( Upper ) , Burrishoole , Costello , Crumlin , Delphi , Erriff , Gowla , Inagh , Inverbeg , Invermore , Kylemore , Newport , and Screebe [16] . The following 32 Welsh rivers served as controls: Aeron , Afan , Arto , Cleddau , Clwyd , Conwy , Dee , Dwyfawr , Dwyryd , Dyfi , Dysynni , Glaslyn , Gwendreath , Gwyrfai , Llyfni , Lougher , Mawddach , Neath , Nevern , Ogmore , Ogwen , Rheidol , Rhymney , Seiont , Taf , Taff , Tawe , Teifi , Tywi , Usk , Wye , and Ystwyth [26 , 27] . Trout caught and released are included in catch data from both countries . Only catch estimates were available for most of these rivers . Recruitment could not be derived , because anadromous brown trout interbreed with freshwater resident trout , about which very few data are available , so this stock was only included in the returns modeling ( not survival ) . Farmed salmon production for all of Ireland was used in modeling [28] , because the majority of farms are in the region where the exposed populations breed . This will tend to have a conservative effect , resulting in a lower estimate of the impact of aquaculture , per tonne of salmon farming . We compared marine plus rod catches of Atlantic salmon from the east coast of Scotland to catches from the west coast of Scotland for the years 1971 to 2004 . Salmon farms appear to be located in the majority of bays on the west coast of Scotland in well over 300 sites ( http://www . marlab . ac . uk/Uploads/Documents/fishprodv9 . pdf ) , so all salmon from rivers on this coast were considered exposed . There is no salmon farming on the east coast , so salmon from east coast rivers were controls . For each coast , a single time series of total catch was used in modeling . Marine catch records were from the International Council for the Exploration of the Sea ( ICES ) Working Group on North Atlantic Salmon [28] and rod catch records were from Fisheries Research Services of Scotland ( J . MacLean , personal communication ) . Rod catches included salmon caught and released . These data were only used in modeling returns . Farmed salmon production for all of Scotland was used in modeling [28] , because regional production data were not available . We also used counts of Atlantic salmon of all ages returning to rivers from 1960–2001 in Scotland from Thorley et al ( 2005 ) [29] . The fish counters are maintained by Fisheries Research Services or by Scottish and Southern Energy plc . There were two exposed populations . One is from the Awe Barrage , which empties into a bay with numerous salmon farms . The other is from the Morar River , which is less than 20 km from the nearest salmon farm , in an area of the coast with many farms [8] . Salmon from the control rivers ( on the east coast ) do not pass by salmon farms in Scotland because of the direction of their migration routes [30] , unless they approach the Norwegian coast . There were ten control populations from the following rivers: Aigas , Beanna , Torr Achilty , Dundreggan , Invergarry , Logie , Westwater , Cluni , Erich , and Pitlo . Farmed salmon production for all of Scotland was used in modeling [28] because regional production data were not available . Estimates of marine survival to one sea winter for hatchery ( and two wild ) Atlantic salmon populations from Ireland and Northern Ireland ( 1980–2004 ) were collected and reported by the ICES Working Group on North Atlantic Salmon [28] . Because only survival estimates are provided , these data were only used in the survival analysis . Salmon from hatcheries on the Screebe , Burrishoole , Delphi , and Bunowen Rivers were considered exposed . Populations from hatcheries on the Shannon , Erne , Lee , Bush , and Corrib Rivers , plus wild populations from the Bush and Corrib Rivers were used as controls . Production data were not available on a regional basis , so national values [28] were apportioned to bays into which exposed rivers empty by assuming that 30% of national production is in the Kilkieren Bay , 10% is in Clew Bay , 5% is in each of Killary Harbour and Ballinakill Bay . These proportions are based on maps of salmon farm locations from the Irish Marine Institute [25] , and they approximately match stock numbers collected by the Central Fisheries Board in the years for which stock numbers are available ( P . Gargan , personal communication ) . Years in which each bay was fallowed were obtained from the Central Fisheries Board ( P . Gargan , personal communication ) , and in these years , the fallowed bays are assigned a production of zero . All exposed rivers empty into bays with salmon farms [25] , while control rivers are at least 55 km away from the nearest farm . Two data sets from Newfoundland were examined—marine survival estimates of wild Atlantic salmon from four rivers from 1987 to 2004 were used in the survival analysis , and grilse returns to 21 rivers from 1986 to 2004 were used in the returns modeling [31] . Salmon farming in Newfoundland is confined to Bay d'Espoir on the south coast [32] ( http://www . fishaq . gov . nl . ca/aquaculture/pdf/aqua_sites . pdf ) . Only the Conne River ( in Bay d'Espoir ) was considered exposed; the Little River ( also in Bay d'Espoir ) was excluded because it has been regularly stocked [31] . The Exploits and Rocky Rivers were also removed from the analysis because of stocking [33] . This left three control rivers for the survival analysis: the Campbellton River , the Northeast Brook ( Trepassey ) , and Western Arm Brook . For the returns analysis , there were 18 control rivers: Campbellton , Crabbes , Fischells , Flat Bay Brook , Highlands , Humber , Lomond , Middle Brook , Middle Barachois , Northeast Brook ( Trepassey ) , Northeast ( Placentia ) , Northwest , Pinchgut Brook , Robinsons , Salmon , Terra Nova ( upper and lower ) , Torrent , and Western Arm Brook . Salmon from control rivers are very unlikely to pass salmon farms because of the direction of their migrations [34] . Farmed salmon production data are from Fisheries and Oceans Canada ( DFO ) Statistical Services [32] . We compared Atlantic salmon returns to six rivers in the Bay of Fundy ( New Brunswick and Nova Scotia , Canada ) to returns to four rivers from other areas of New Brunswick and Nova Scotia . We grouped the six exposed rivers into three groups and estimated the impact of aquaculture on each group separately , because salmon from these three groups have different degrees of exposure to salmon farming . The three groups of exposed rivers are the inner Bay of Fundy group ( Stewiacke and Big Salmon Rivers ) , the Saint John River group ( Saint John and Nashwaak Rivers ) , and the outer Bay of Fundy group ( St . Croix and Magaguadavic Rivers ) . Salmon farming in New Brunswick is highly concentrated in the Quoddy region of the outer Bay of Fundy ( http://www . gnb . ca/0177/10/Fundy . pdf ) , although some farms are also found along the Nova Scotia coast of the Bay of Fundy . Salmon from control rivers enter into the Atlantic directly ( LaHave River ) or into the Gulf of St . Lawrence ( Restigouche River , Miramichi River , Catamaran Brook ) and do not pass by farms during their migrations . The same controls are used for all comparisons in New Brunswick and Nova Scotia . The estimates of returns to the rivers are published by DFO [28 , 35–40] . Outer Bay of Fundy salmon must pass through an area containing many salmon farms early during their migrations [41] . Although Saint John River salmon enter the ocean in an area without salmon farms , they are known to pass through the region containing many farms early during their migrations [41] . Salmon from inner Bay of Fundy rivers are considered exposed to salmon farming despite being up to 260 km away because of historical information indicating that juvenile salmon from these populations are found during the summer and fall in the area where salmon farms are currently located [42] . However , the evidence that this region is important habitat for inner Bay of Fundy and Saint John River populations is mixed [43] . For this reason , we ran an alternative model with only outer Bay of Fundy populations considered exposed , and all other New Brunswick and Nova Scotia rivers as controls . For all New Brunswick rivers , an estimate of egg deposition was used as an index of spawners , to account for a significant increase in the age of spawners in many rivers over the study period . The number of grilse ( salmon maturing after one winter at sea ) and large spawners ( repeat spawners or salmon maturing after two or three winters at sea ) in each year was multiplied by a river-specific estimate of fecundity for a salmon of that size . Then , the index of spawners in a given year was derived by adding up all the eggs that could produce smolts in a year y , using river-specific ages at smolting from the literature . Returning hatchery-origin spawners are also added to the “spawners” but not to “returns . ” “Recruits” is the number of grilse that return to each river in year y + 1 , so that ( in Equation 1 ) is the number of grilse returning per egg that would have smolted in year y . Estimates of returns to rivers from traps and other surveys were used in the returns analysis . No corrections were made to account for marine fisheries , but marine exploitation has been quite limited since the late 1980s , when salmon farming became a substantial industry [44] . Farmed salmon production data are from DFO Statistical Services [32] . For coho salmon in British Columbia ( BC ) , spawner estimates are based on DFO's escapement database ( NuSEDS ) , which includes estimates of spawning salmon of all species for hundreds of rivers and streams on the BC coast since 1950 ( P . VanWill , DFO Pacific , unpublished data ) . We considered rivers on the east side of the Queen Charlotte and Johnstone Straits to be exposed ( all rivers from Wakeman Sound to Bute Inlet , DFO Statistical Areas [SAs] 12 and 13 ) . All rivers on the BC Central Coast from Finlayson Channel to Smith Inlet ( SAs 7 , 8 , 9 , and 10 ) were included as controls . In the regions considered exposed in BC , all salmon must pass by farms to get into the open ocean , although in some cases , the farms are at the end of long channels down which the salmon migrate ( as far as 90 km in the most extreme case ) . Control populations to the north do not pass by farms , because of the direction of their migration routes [45] . Coverage in the NuSEDS database varies considerably in time and space , as does the quality of the estimates . We changed all indicators of unknown values ( including “none observed” and “adults present” ) to a common missing value indicator . To reduce effects of inconsistent monitoring procedures , only data since 1970 were included in the analysis . All rivers known to be regularly stocked with hatchery salmon or to contain constructed spawning channels were also removed from exposed and control areas , leaving 49 exposed and 70 control rivers . Estimates were combined for each SA , the smallest areas for which catch rates are estimated . This was done by modeling returns to each SA and year , using a generalized linear model with negative binomial errors . The predicted returns for each SA were then used as spawner estimates ( Si , y in Equation 1 ) . To derive recruitment estimates , we followed Simpson et al . ( 2004 ) [46] , applying exploitation rate estimates from Toboggan Creek ( J . Sawada , DFO Pacific , personal communication ) to the controls , and the average of the exploitation rates for Quinsam Hatchery , Big Qualicum Hatchery , and the Black Creek wild indicator population to the exposed stocks . After 1998 , only the estimates from Black Creek were used for exposed stocks . Recruitment estimates for coho were based on the assumption that coho follow a fixed 3-y life cycle . For pink , chum , and coho salmon , aquaculture production estimates include all salmon species farmed in SAs 12 and 13 ( the Queen Charlotte and Johnstone Straits ) from 1990 to 2003 ( H . Russell , BC Ministry of Agriculture , Food , and Fisheries , unpublished data ) . In years when two or fewer companies were raising salmon in either area , estimates were not available . BC salmon farm locations are made available at http://www . al . gov . bc . ca/fisheries/licences/MFF_Sites_Current . htm . Estimates of pink salmon spawner abundance were derived in the same manner as described above for coho salmon . “Returns” are spawners plus catch for a given year , assuming a fixed two year life cycle . The same regions were considered exposed , but because enumeration varies by species , there were only 36 exposed rivers from SAs 12 and 13 ( from Wakeman Sound to Bute Inlet ) included . Wood et al . ( 1999 ) [47] consider the pink salmon catches in SAs 8 , 9 , and 10 to consist mainly of salmon returning to those areas ( respectively ) , so catch data from DFO [48] were used in each of these SAs . Area 7 was excluded from the survival analysis because catches for SA 7 are difficult to estimate due to the adjacent regions being much larger [47] , leaving 47 control rivers from Burke Channel to Smith Inlet . For Queen Charlotte and Johnstone Straits ( the exposed areas ) , DFO does not estimate catches at the level of individual SA . To obtain approximate returns to each exposed SA , we found the proportion of total escapement to the Straits that was in our dataset ( i . e . , regularly enumerated rivers on the east side of the Straits without a major hatchery or constructed spawning channel ) and assumed the same proportion of the total catch would be returning to those rivers ( i . e . , assumed equal catchability across stocks ) . For odd years , we used estimates from the Pacific Salmon Commission ( B . White , unpublished data ) of the catch of pink salmon in Johnstone and Georgia Straits that were not returning to the Fraser River . In even years , there is no pink salmon run on the Fraser River , so total returns to the Straits could be used . For chum salmon , we used estimates of returns ( i . e . , before exploitation ) and spawners to large coastal areas [49] . Chum from the east side of Queen Charlotte and Johnstone Straits , from Wakeman Sound to Bute Inlet ( SAs 12 and 13 ) were considered exposed to salmon farming , while chum from the Central Coast from Bute Channel to Seymour Inlet ( SAs 8–11 ) were considered controls . Estimates were available as a single time series for the exposed area , and a time series for each SA for the controls . An index of recruits per spawner was generated by lining up returns with spawners according to age distributions given in Ryall et al . ( 1999 ) [50] , to 1998 , and then the average values from 1988–1998 for the subsequent years , to 2003 . | The impact of salmon farming on wild salmon and trout is a hotly debated issue in all countries where salmon farms and wild salmon coexist . Studies have clearly shown that escaped farm salmon breed with wild populations to the detriment of the wild stocks , and that diseases and parasites are passed from farm to wild salmon . An understanding of the importance of these impacts at the population level , however , has been lacking . In this study , we used existing data on salmon populations to compare survival of salmon and trout that swim past salmon farms early in their life cycle with the survival of nearby populations that are not exposed to salmon farms . We have detected a significant decline in survival of populations that are exposed to salmon farms , correlated with the increase in farmed salmon production in five regions . Combining the regional estimates statistically , we find a reduction in survival or abundance of wild populations of more than 50% per generation on average , associated with salmon farming . Many of the salmon populations we investigated are at dramatically reduced abundance , and reducing threats to them is necessary for their survival . Reducing impacts of salmon farming on wild salmon should be a high priority . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"ecology"
] | 2008 | A Global Assessment of Salmon Aquaculture Impacts on Wild Salmonids |
Following productive , lytic infection in epithelia , herpes simplex virus type 1 ( HSV-1 ) establishes a lifelong latent infection in sensory neurons that is interrupted by episodes of reactivation . In order to better understand what triggers this lytic/latent decision in neurons , we set up an organotypic model based on chicken embryonic trigeminal ganglia explants ( TGEs ) in a double chamber system . Adding HSV-1 to the ganglion compartment ( GC ) resulted in a productive infection in the explants . By contrast , selective application of the virus to distal axons led to a largely nonproductive infection that was characterized by the poor expression of lytic genes and the presence of high levels of the 2 . 0-kb major latency-associated transcript ( LAT ) RNA . Treatment of the explants with the immediate-early ( IE ) gene transcriptional inducer hexamethylene bisacetamide , and simultaneous co-infection of the GC with HSV-1 , herpes simplex virus type 2 ( HSV-2 ) or pseudorabies virus ( PrV ) helper virus significantly enhanced the ability of HSV-1 to productively infect sensory neurons upon axonal entry . Helper-virus-induced transactivation of HSV-1 IE gene expression in axonally-infected TGEs in the absence of de novo protein synthesis was dependent on the presence of functional tegument protein VP16 in HSV-1 helper virus particles . After the establishment of a LAT-positive silent infection in TGEs , HSV-1 was refractory to transactivation by superinfection of the GC with HSV-1 but not with HSV-2 and PrV helper virus . In conclusion , the site of entry appears to be a critical determinant in the lytic/latent decision in sensory neurons . HSV-1 entry into distal axons results in an insufficient transactivation of IE gene expression and favors the establishment of a nonproductive , silent infection in trigeminal neurons .
Herpes simplex virus type 1 ( HSV-1 ) and 2 ( HSV-2 ) are prototypic members of the genus Simplexvirus within the herpesvirus subfamily Alphaherpesvirinae . In vitro , HSVs are pantropic , causing lytic infections in various tissues and cell types of a broad range of host species [1] . In vivo , humans are the only natural hosts of HSVs , and infection is almost exclusively limited to the epithelial cells and neurons of the peripheral nervous system ( PNS ) . The portal of entry in HSV-1 infections is the oronasal mucosa , where the virus spreads rapidly with productive , lytic infection of epithelial cells [2] . HSV-1 reaches the PNS by entry into free nerve endings that are in contact with the infected epithelium , and by retrograde axonal transport [3] . Beyond the neonatal period , replication of HSV-1 within the PNS is tightly controlled , and further ascending spread into the central nervous system is prevented in the immunocompetent host . As a result , HSV-1 establishes latency in surviving neurons , converting them into a lifelong reservoir of recurrent infection , which occurs in response to diverse stimuli causing neuronal stress [4] . The ability to switch from rapidly progressing , lytic spread in epithelia to a nonproductive , latent infection in sensory neurons is fundamental to the life cycle of HSVs and other related alphaherpesviruses . Although crucial for our understanding of the pathogenesis of alphaherpesvirus infection , the events that ultimately trigger the establishment of latent infection in sensory neurons are not fully understood . The latency-associated transcript ( LAT ) is abundantly expressed in infected neurons , and has been shown to promote the establishment and maintenance of latency , in part because of the anti-apoptosis functions of LAT and the ability of micro-RNAs and other small non-coding RNAs encoded by LAT to interfere with productive infection [4] . The viral proteins ICP0 and VP16 have both been implicated in the lytic/latent decision in the infected neuron [5] . The HSV-1 immediate-early ( IE ) regulatory protein ICP0 is a RING finger E3 ubiquitin ligase that acts as a promiscuous activator of gene expression and is critical for the efficient initiation of productive infection , the prevention of cellular silencing of viral transcription , and reactivation from latency [6] . Central to the function of HSV-1 ICP0 and functional homologues in other alphaherpesviruses is the ability of ICP0 to degrade PML nuclear bodies ( also known as nuclear domain 10 , ND10 ) and to disrupt the ND10-dependent antiviral interferon response and other cellular functions linked to ND10 [7] , [8] . HSV-1 ICP0 has been found to functionally interact with class II histone deacetylases ( HDAC ) , to dissociate HDAC from the lysine-specific demethylase 1/REST/CoREST repressor complex , and to promote histone removal and acetylation , thus preventing the formation of inactive chromatin on the HSV-1 genome [9]–[11] . Posttranslational histone modifications have been implied in HSV gene expression during lytic and latent infection , and the establishment , maintenance , and reactivation from latency [12]–[16] . HSV-1 VP16 ( also designated pUL48 or alpha-transinducing factor , αTIF ) is a structural component of the tegument layer . Upon fusion with the cytoplasmic membrane of monolayer cells , HSV-1 particles deliver 500–1 , 000 molecules of VP16 into the cytoplasm , which are transported to the nucleus independently of capsids [17] . Although purified in vitro de-enveloped HSV-1 particles containing a VP16-EGFP fusion protein were reported to move in a retrograde direction along microtubules when injected into squid giant axons [18] , several studies of HSV-1 and other alphaherpesviruses have demonstrated that VP16 dissociates from viral particles upon entry into the host cell and that capsids are transported to the nucleus independently of VP16 [19]–[21] . Live-cell imaging experiments examining the retrograde axonal transport of pseudorabies virus ( PrV ) and HSV-1 in neurons of human , mouse and avian origin have shown that VP16 and other proteins of the outer tegument layer are predominantly lost from the nucleocapsid prior to the onset of retrograde axonal transport , and do not move with the capsid to the nucleus [22] . However , it was also noted that to some extent VP16 appears to be axonally transported in retrograde direction independent of capsids . In lytic infection , VP16 forms a tripartite complex with the cellular proteins HCF-1 and Oct-1 , which binds to the TAATGARAT elements present in HSV IE promoters and acts as a potent transcriptional activator of IE gene expression [23]–[26] . The transcriptional activation domain of HSV-1 VP16 ( VP16AD ) interacts with a large number of cellular factors that are involved in gene activation [27] . Although not essential for IE gene expression , coactivators recruited by the HSV-1 VP16AD contribute to relatively low levels of histones on the viral genome during lytic infection [28]–[31] . VP16 is essential for stress-induced HSV-1 reactivation in vitro [32] . Exit from latency following heat shock in the mouse ocular model has been reported to depend on the de novo activation of the VP16 promoter and synthesis of VP16 in infected neurons [33] . In stressed neurons , HCF-1 has been shown to relocalize from the cytoplasm to the nucleus and to be recruited to HSV-1 IE promoters [34] . The regulated relocalization of de novo synthesized VP16 and HCF-1 from the cytoplasm to the nucleus of stressed neurons appears to be a critical step in the initiation of lytic gene expression during reactivation from latency [35] . In addition to its regulatory function in IE gene expression , VP16 and homologous alphaherpesvirus proteins of the outer tegument layer mediate essential functions related to viral egress [36] . At present , animal models allow only a pinpoint , snapshot-like observation of the critical early phase of viral arrival in the PNS and onset of replication . Furthermore , there is enormous variation in the outcome of HSV-1 infection of the nervous system in laboratory animals . In mice , the course of infection depends on various factors , including the viral strain , infectious dose , route of infection , mouse strain and age , and prior immunization history of the animal [37] . As an alternative to animal models , classical or modified Campenot chambers and microfluidic devices have been used to study the directional spread of alphaherpesviruses between epithelial cells and neurons in compartmentalized organotypic culture systems [38]–[42] . The directional spread of HSV-1 between epithelia and the PNS has been analyzed in our laboratory by establishing compartment cultures based on chick embryonic corneal epithelial cells and trigeminal ganglion ( TG ) explants ( TGEs ) [43] . The aim of the present study was to identify factors triggering either productive or silent infection of sensory neurons by mimicking the critical early phase of primary PNS infection in organotypic cultures ex vivo . Interestingly , remarkable differences in the ability of HSV-1 to initiate productive infection in sensory neurons were observed depending on the site used to infect the explants . In accordance with earlier reports [22] , [43] , [44] , productive infection was easily induced in embryonic chicken neurons by directly adding the viral inoculum to the explant cultures . In contrast , upon entry into distal axons , incoming HSV-1 genomes were largely destined to a quiescent , latency-like infection , characterized by the expression of high levels of the latency-associated transcript ( LAT ) . HSV-1 remained responsive to the IE gene transcriptional inducer hexamethylene bisacetamide ( HMBA ) , and to transcriptional transactivation by simultaneous co-infection of the ganglion compartment ( GC ) with helper virus . Efficient transactivation was dependent on the presence of functional VP16 in HSV-1 helper virus particles . To our knowledge , this is the first report to provide experimental evidence that axonal infection and preceding retrograde axonal transport of subviral particles strongly diminishes the ability of HSV-1 to initiate a productive infection of neurons , most likely via a failure of VP16-induced expression of lytic genes .
In the present study we analyzed the course of a synchronous HSV-1 infection of TGE cultures resulting from the application of a defined viral inoculum to either free nerve endings or directly to the explants . TGEs were derived from day-15 chicken embryos , as described previously [43] . Selective HSV-1 infection of distal axons was achieved with the aid of a Campenot-like compartment chamber [38] , in which the inner compartment containing the TGE ( i . e . , the GC ) was separated by a leak-proof diffusion barrier from the outer compartment containing the distal axons [the axonal compartment ( AC ) ] ( Figure 1A ) . Infection of the respective compartment was performed after 5–6 days of in vitro cultivation . Staining of the AC with the retrograde neuronal tracer DiI showed that at this time point approximately 200 neurites/TGE had crossed the diffusion barrier and reached the AC . Neurites within the AC were essentially free of accompanying glial cells . By 48 h after applying DiI to the AC , approximately 100 neurons/TGE could be visualized in intact cultures . DiI-positive neurons exhibited a characteristic morphology with a large , rounded soma , and were mostly localized at the margin of the explant . Many neurons contained prominently stained neurites ( Figure 1B–D ) . Direct infection of the GC with HSV-1 17 CMV-IEproEGFP expressing EGFP under control of the CMV IE-promoter , and monitoring of the reporter protein expression in intact cultures by fluorescence microscopy demonstrated that embryonic chicken TGEs are susceptible to productive HSV-1 infection under the culture conditions used . By 24 h after infection [hours post infection ( hpi ) ] of the GC with 1×106 plaque-forming units ( pfu ) of HSV-1 17 CMV-IEproEGFP , the vast majority of cells directly accessible to the virus inoculum strongly expressed EGFP . Costaining of the AC with DiI allowed the identification of EGFP-positive neurons with neurites reaching the AC ( Figure 2A ) . Approximately 30% of DiI-positive neurons ( 53 out of 163 neurons ) were found to be EGFP-positive at 24 hpi ( Figure 2C ) . According to their morphology , most of the remaining DiI-negative , EGFP-positive cells appeared to be macroglial cells , most likely Schwann cells . We next studied the retrograde spread of HSV-1 in intact TGEs by the addition of HSV-1 17 CMV-IEproEGFP to the AC and daily monitoring of cultures by fluorescence microscopy . In preliminary experiments , various infectious doses and protocols for infection of the AC were evaluated . Except where indicated otherwise , an inoculum of 5×106 pfu of HSV-1 added for 1 h to the AC was used in all subsequent experiments . This infectious dose corresponded to a multiplicity of infection ( MOI ) of approximately 5 in a monolayer culture in a 35-mm-diameter dish , and did not exhibit unwanted stimulatory effects on the TGEs due to serum components present in the virus stocks . Replacement of the TG medium in the AC by serum-containing media stimulated the outgrowth of glial cells or fibroblasts into the AC and occasionally led to the contraction and partial detachment of the explants ( data not shown ) . As shown in Figure 2B , the diffusion barrier proved to be tight , effectively inhibiting the nonspecific infection of cells that surrounded the explant , and were in direct contact with the inner rim of the glass cylinder . However , EGFP expression in the TGEs was unexpectedly poor . At 24 hpi , less than half of the axonally-infected cultures contained single EGFP-positive cells . Staining of the AC with DiI immediately after infection showed that EGFP expression was restricted to just a few isolated neurons ( Figure 2B ) . Less than 2% of DiI-positive neurons ( 2 out of 127 neurons ) were found to be EGFP-positive ( Figure 2C ) . The low number of reporter protein expressing neurons in cultures axonally-infected with HSV-1 17 CMV-IEproEGFP at 24 hpi prompted us to monitor the kinetics of EGFP expression in intact TGEs over a longer time period ( Figure 2D–F ) . During the first 10 days after infection [days post infection ( dpi ) ] , the percentage of cultures containing EGFP-positive cells slowly increased from approximately 40% on 1 dpi to approximately 70% between 4 and 6 dpi , and decreased thereafter . Between 2 and 8 dpi , small plaque-like clusters of EGFP-positive cells appeared in approximately 15% of cultures , persisted for a few days , and disappeared thereafter , indicating a self-limiting secondary spread of HSV-1 within the TGEs ( Figure 2D , F ) . In axonally infected TGEs , cell free virus could not be isolated from the supernatants of the GC ( data not shown ) . The long-term monitoring of these cultures at weekly intervals showed that in 10–20% of cultures , low and fluctuating levels of EGFP-expressing neurons persisted up to 10 weeks postinfection ( wpi ) ; in one culture a plaque-like cluster of infected cells developed at 9 wpi ( Figure 2D ) . The onset of EGFP expression in individual axonally-infected neurons was followed up with the aid of serial photographic documentation . This demonstrated that the number of EGFP-expressing neurons peaked at 4 dpi , and then decreased to low and fairly stable levels thereafter ( Figure 2E ) . Determination of the cumulative number of EGFP-positive neurons revealed differences in the kinetics of the onset of EGFP expression , with an approximately fivefold increase in the cumulative number of positive neurons between 1 and 7 dpi , and a long-lasting but minimal ( i . e . , less than 1 . 1-fold ) increase between 3 and 10 wpi ( Figure 2E ) . As demonstrated in Figure 2G and H , the spread of the parental wild-type ( wt ) strain HSV-1 17syn+ closely resembled the pattern of fluorescence observed in HSV-1 17 CMV-IEproEGFP-infected TGEs . Two days after infection of the AC , single cells or small plaque-like clusters of HSV-1-infected cells located in the interior of explants could be detected in a few TGEs ( Figure 2G ) . In contrast , after direct infection of the GC with 104 pfu of HSV-1 17syn+ , spread of HSV-1 into the interior parts of TGEs was observed , and large areas of HSV-1-positive cells containing infected glial cells and neurons could be visualized ( Figure 2H ) . To further characterize the replication of HSV-1 in TGEs , we determined viral genome and transcript levels in infected TGEs by quantitative real-time PCR ( qPCR ) . Infection of the GC with HSV-1 17syn+ resulted in the onset of a productive infection of the TGEs . As compared to 1 hpi ( input genomes ) , the number of replicated genomes increased exponentially between 12 and 48 hpi ( Figure 3A ) . The viral genome and transcript levels in axonally infected TGEs were analyzed in groups of ten cultures , containing two explants each . At 6 , 24 , and 168 hpi with 5×106 pfu of HSV-1 17syn+ added to the AC , the median HSV-1 DNA levels in the explants were approximately 14 , 000 , 17 , 000 , and 21 , 000 genomes/TGE , respectively ( Figure 3B ) . Although the HSV-1 DNA load in individual TGEs varied markedly and the variations increased with time post infection , there were no significant differences in viral genome levels in the GC between 6 hpi and 168 hpi ( i . e . 7 dpi ) . The mode of entry into distal axons and the specificity of this qPCR-based experimental approach were tested by inhibition of the microtubule-mediated axonal transport with nocodazole and application of fusion-deficient virus particles to the AC . The addition of 10 µM nocodazole to both compartments simultaneously with viral infection of the AC led to a highly significant reduction in viral DNA levels in the explants at 2 hpi ( Figure 3C ) . Negative effects of the nocodazol treatment on neurites within the AC were not visible by light microscopy ( data not shown ) . Infection of the AC with 1×108 particles of the gH-negative mutant HSV-1 KOSgH87 purified from the supernatants of transcomplementing gH-expressing Vero cells ( Vero F6gH ) , corresponding to 5×106 pfu on VeroF6gH cells , resulted in similar genome levels in the GC at 6 hpi to those observed in TGEs infected with HSV-1 17 syn+ . In contrast , genomes could not be detected in TGEs after the addition of an identical number of particles of HSV-1 KOSgH87 purified from the supernatants of nontranscomplementing Vero cells ( corresponding to 5 pfu on VeroF6gH cells ) to the AC ( Figure 3D ) . Genome levels and the number of primary infected cells in TGEs were quantified in preparations of dispersed TGEs infected in the GC and the AC , respectively , in the presence of 50 µg/ml aciclovir ( ACV ) and harvested at 2 dpi . After infection of the AC with 1×106 pfu of HSV-1 17syn+ , approximately 2 , 000 HSV-1 genomes/TGE were present in the dispersed cultures . HSV-1 antigen-positive neurons or nonneuronal cells could not be detected by immunofluorescence . In contrast , infection of the GC led to the dose-dependent expression of viral antigens in neurons and nonneuronal cells ( Table 1 ) . Absolute and relative levels of HSV-1 IE ( ICP27/UL54 ) , early/late ( E/L; gB/UL27 ) , and late ( L; gC/UL44 ) transcripts ranged from 500 to 125 , 000 copies/TGE and 1 . 8×10−6 to 6×10−4 ( viral transcripts/ß-actin transcripts ) , respectively , at 6 , 24 and 168 hpi ( Figure 3E ) . IE , E/L , and L transcript levels close to the detection limit of reverse transcription ( RT ) -PCR were detectable in few cultures at 6 hpi , while thereafter there were moderate but significant increases in lytic transcript levels . Lytic transcripts were present in most cultures at 24 hpi and in approximately half of the cultures at 168 hpi . However , the onset of lytic gene expression in axonally-infected TGEs was remarkably low relative to the high genome levels present in the GC . In contrast to lytic transcripts , high levels of the 2 . 0-kb major LAT transcript were present in all TGEs selectively infected via distal axons at 24 and 168 hpi . Between 24 and 168 hpi , LAT levels increased by approximately 40-fold . The median level of LAT at 168 hpi was approximately 1 , 000-fold higher than that in ICP27 transcripts ( Figure 3E ) . LAT was also expressed in cultures infected in the GC with 5×106 to 8×103 pfu of HSV-1 17syn+ . As compared to TGEs infected in the AC , the specific transcriptional activity of the LAT gene ( transcripts/genome ) was 350 to 8 , 000 -fold lower in TGEs infected in the GC ( Figure 3F ) . To exclude strain- and type-specific effects on HSV growth in TGEs after axonal infection , we repeated the experiments with clinical isolates of HSV-1 and HSV-2 . Again , a synchronous productive infection with exponentially rising genome levels between 1 and 3 dpi was not induced in TGEs infected in the AC ( data not shown ) . Since entry of HSV-1 into the distal axons of sensory neurons led to a predominantly nonproductive infection , we tested whether productive infection could be induced by treatment with HMBA , a known stimulator of HSV-1 IE gene expression [45] . The addition of 2 . 5 mM HMBA to the culture medium of axonally-infected TGEs significantly increased the number of neurons expressing EGFP under control of the CMV IE promoter and the HSV-1 gD promoter at 24 hpi ( Figure 4A ) . Quantification of HSV-1 genome and transcript levels indicated that HMBA treatment induced viral genome replication in axonally-infected TGEs and significantly increased the expression of lytic genes at 24 hpi but did not affect the expression of LAT ( Figure 4B , C ) . At 7 dpi , the median viral genome levels in HMBA-treated cultures were approximately 1 , 000-fold higher than in controls ( Figure 4C ) . Monitoring of intact cultures demonstrated that HMBA strongly induced productive HSV-1 infection in non-neuronal cells . At 8 dpi , approximately 40% of the explants showed signs of massive viral spread ( Figure 4D ) . When HMBA was added at 7 dpi to TGEs axonally infected with HSV-1 17 CMV-IEproEGFP , no effect on the number of EGFP-expressing cells was observed ( data not shown ) . In order to analyze the effect of HMBA on the release of cell free infectious virus into the culture supernatants , we infected TGEs with 1×104 pfu HSV-1 17 gDproEGFP in the GC using culture media without CMC and HSV-1 antiserum . As compared to nontreated explants , HMBA-treatment led to an approx . 30-fold increase in the median levels of cell free infectious virus in the supernatants at 4 dpi . The number of infectious particles released by repeated freeze-thawing of explants was approximately 10-fold higher in HMBA-treated cultures ( Figure 4E ) . We also tested whether genome replication and the expression of lytic genes in axonally-infected neurons could be induced by simultaneous co-infection of the GC with a helper virus . To this end we established a method for quantifying HSV-1 helper-virus-induced transcriptional transactivation and genome replication after the selective entry of HSV-1 reporter virus via distal axons . Cultures were infected in the AC with replication-competent EGFP-expressing HSV-1 mutants , whereas the GC was incubated with the spread-deficient gH-negative mutant HSV-1 KOS gH87 . The use of a spread-deficient , gH-negative HSV-1 helper virus limited transactivation to those cells primarily infected by the helper virus and allowed to us to specifically follow up the replication and reporter gene expression of HSV-1 genomes present in sensory neurons after entry within the AC . In contrast to infection of the AC with 5×106 pfu of HSV-1 17 CMV-IEproEGFP only ( see above ) , EGFP expression in TGEs was reliably induced by simultaneous co-infection of the GC with 5×106 pfu of the helper virus HSV-1 KOS gH87 . Staining of the AC with DiI immediately after infection with HSV-1 demonstrated that at 24 hpi , most EGFP-expressing cells exhibited a typical neuronal morphology and were stained by DiI ( Figure 5A ) , indicating the occurrence of transactivation of silent HSV-1 infection in neurons . The number of EGFP-expressing cells was approximately 30-fold higher in TGEs co-infected with HSV-1 helper virus ( Figure 5B ) . EGFP DNA and RNA levels were determined by qPCR to quantify the transactivation of the reporter viruses by the HSV-1 helper virus . The simultaneous co-infection of the GC with 5×106 pfu of HSV-1 KOS gH87 led to an approximately 20-fold increase in the genome levels of HSV-1 17 CMV-IEproEGFP in axonally-infected TGEs at 24 hpi ( Figure 5C ) . Serial fivefold dilution of the helper virus revealed that a significant induction of reporter virus replication was still achieved by adding 2×105 pfu helper virus to the GC . To estimate the ratio of helper to reporter virus needed for transactivation , helper virus genome levels resulting from co-infection of the GC were quantified . Approximately 40 , 000 HSV-1 genomes/TGE were detected by qPCR in cultures immediately harvested after infection of the GC with 2×105 pfu of HSV-1 KOS gH87 helper virus ( i . e . , the lowest amount of helper virus able to significantly transactivate reporter virus; see above ) . Comparable to what was observed with HSV-1 17syn+ , infection of the AC with 5×106 pfu of HSV-1 17 CMV-IEproEGFP reporter virus resulted in mean levels of approximately 10 , 000 HSV-1 genomes/TGE at 6 hpi . This corresponds to an overall ratio of helper to reporter virus genomes in co-infected TGEs of approximately 4 . Considering that the HSV-1 added to the GC also infects nonneuronal cells ( Figure 2A ) , that only a subfraction of axonally-infected neurons may be directly accessible to the spread-deficient helper virus in en-bloc-cultivated TGEs , and that the genome/pfu ratio of the helper virus preparation used was approximately 30 , the actual ratio of helper to reporter virus needed to transactivate HSV-1 in axonally-infected neurons is probably much lower . Quantification of transcript levels showed that simultaneous co-infection of the GC with HSV-1 helper virus strongly increased EGFP transcript levels at 24 hpi with HSV-1 17 CMV-IEproEGFP and HSV-1 17 gDproEGFP ( Figure 5D ) . As observed for the induction of genome replication , at least 2×105 pfu of helper virus needed to be applied to the GC to achieve a significant transactivation of EGFP expression in cultures axonally-infected with HSV-1 17 CMV-IEproEGFP ( data not shown ) . In order to further elucidate the mechanism underlying helper-virus-induced HSV-1 IE gene transactivation in axonally-infected TGEs , reporter gene expression in HSV-1-IE4proEGFP-infected cultures was determined in the absence of de novo protein synthesis at 6 hpi . As shown in Figure 5E , low levels of EGFP expression were detected in most of the cycloheximide ( CHX ) -treated cultures without helper virus . The simultaneous addition of 5×106 pfu of helper virus significantly increased the EGFP expression , indicating that transactivation of reporter gene expression under the control of the HSV-1 ICP4 promoter occurred directly ( i . e . , by incoming helper virus particles and in the absence of de novo protein synthesis ) . We also investigated whether the addition of HSV-1 helper virus to axonally-infected cultures was able to transactivate reporter virus genomes at 7 dpi ( i . e . , after the establishment of an LAT-positive , silent infection ) . These experiments showed that the HSV-1 reporter virus was refractory to transactivation by HSV-1 helper virus at 7 dpi . Infection of the GC with HSV-1 KOSgH87 did not significantly increase reporter virus genomes or gene expression in explants axonally-infected with HSV-1 17 CMV-IEproEGFP ( Figure 5F , G ) , or directly transactivate IE gene expression in HSV-1-17-IE4proEGFP-infected cultures in the presence of CHX ( Figure 5H ) . The transactivation of the ICP4 promoter in axonally-infected and CHX-treated cultures by simultaneous infection of the GC with HSV-1 helper virus suggested a direct VP16-dependent transcriptional activation of IE gene expression . To study the role of VP16 , we repeated the experiments using the VP16AD-negative mutant HSV-1 KOS RP5 as a helper virus to infect the GC . The corresponding revertant HSV-1 KOS RP5R served as a control [46] . The number of particles of HSV-1 KOS RP5 used to infect the GC was adjusted to the number of particles present in 5×106 pfu of the HSV-1 KOS RP5R preparation . Following axonal infection of TGEs with HSV-1 IE4proEGFP in the presence of CHX , IE gene expression was significantly increased by infection of the GC with HSV-1 KOS RP5R , whereas the addition of RP5 did not lead to a significant transcriptional transactivation of the reporter gene at 6 hpi ( Figure 6A ) . When added to the GC in the absence of CHX at a concentration of 108 particles , both HSV-1 KOS RP5 and RP5R were able to significantly transactivate IE gene expression of the reporter virus at 6 hpi ( Figure 6B ) . In agreement with this finding , a significant increase of reporter virus genomes was observed at 24 hpi after the addition of 108 particles of HSV-1 KOS RP5 and RP5R helper viruses . Serial ten-fold dilution of the helper viruses demonstrated that the efficiency of transactivation by the rescue mutant HSV-1 KOS RP5R was 10 to 100-fold higher as compared to RP5 ( Figure 6C ) . The ability of HSV-1 KOS RP5 to infect distal axons was also analyzed . Similar to what was observed in TGEs at 24 hpi with HSV-1 17syn+ , only basal levels of lytic transcripts were present in TGEs axonally-infected with HSV-1 KOS RP5 , whereas LAT was expressed at high levels ( data not shown ) . In CHX-treated , axonally infected TGEs we could not detect significant VP16-dependent differences in the transcriptional activity of ICP27 between HSV-1 KOS RP5 and RP5R at 6 hpi ( Figure 6D ) . In contrast , the transcriptional activity of ICP27 in CHX-treated TGEs infected in the GC was approximately two log-orders higher in HSV-1 KOS RP5R-infected cultures as compared to HSV-1 KOS RP5 ( Figure 6D ) . Similar differences between HSV-1 KOS RP5 and RP5R were observed in CHX-treated Vero cell monolayers at 6 hpi ( Figure 6D ) . As observed with HSV-1 17syn+ , HMBA-treatment led to a significant increase in genome levels in TGEs axonally infected with HSV-1 KOS RP5 at 24 hpi ( Figure 6E ) . The HSV-1-independent transactivation of HSV-1 IE gene expression by co-infection of the GC with the alphaherpesviruses HSV-2 and PrV was investigated using the HSV-2 strain 333 and the replication-competent PrV mutant PrV-KaΔgGgfp . As observed with HSV-1 , direct addition of HSV-2 to the GC resulted in exponentially rising genome levels indicative of massive viral replication in the TGEs , whereas addition of HSV-2 to the AC did not induce productive infection of the TGEs between 12 hpi and 48 hpi ( Figure 7A ) . The simultaneous addition of 5×106 pfu of HSV-2 helper virus to the GC of TGEs axonally-infected with HSV-1 17 CMV-IEproEGFP led to highly significant increases in HSV-1 genomes and EGFP transcript levels at 24 hpi ( Figure 7 B , C ) . In contrast to the HSV-1 helper virus , infection of the GC with the HSV-2 helper virus at 7 dpi of the AC with HSV-1 17 CMV-IEproEGFP also strongly increased HSV-1 genomes and EGFP transcripts ( Figure 7B , C ) . In the absence of de novo protein synthesis , HSV-2 helper virus strongly transactivated EGFP expression under control of the HSV-1 ICP4 promoter ( Figure 7 D ) . TGEs were found to be highly susceptible to direct infection of the GC with PrV . Furthermore , in contrast to HSV-1 and HSV-2 , PrV infection of the AC also led to a rapid , productive spread of PrV in TGEs ( Figure 8A ) . PrV-KaΔgGgfp proved to be a potent transactivator of HSV-1 IE gene expression in infected neurons . The simultaneous addition of 5×106 pfu of PrV to the GC led to a highly significant increase in HSV-1 ICP27 gene expression at 6 hpi ( Figure 8B ) . As observed with HSV-2 and in contrast to HSV-1 helper virus , infection of the GC with PrV helper virus at 7 dpi of the AC with HSV-1 significantly increased HSV-1 IE gene expression ( Figure 8B ) . Simultaneous co-infection of the GC with helper virus in the presence of CHX resulted in a significant increase in the basal level of HSV-1 IE gene expression ( Figure 8C ) . Additionally , we studied transactivation by PrV helper virus in TGEs co-infected in the AC with HSV-1 and PrV . Axonal co-infection of the TGEs resulted in a strong transactivation of HSV-1 IE gene expression by PrV helper virus at 6 hpi . Transactivation was abrogated , however , by CHX-treatment of TGEs ( Figure 8D ) . Thus , in axonally co-infected cultures transactivation of HSV-1 appears to be dependent on de novo protein synthesis of PrV .
HSV-1 latency and reactivation in vivo result from a complex interaction between the virus and its host . It is assumed that immediately after virus entry into the host cell , the tegument protein VP16 plays an essential role in the combinatorial regulatory circuit controlling the lytic/latent decision . If the transactivation of IE genes by the VP16-induced complex is insufficient , this may diminish lytic gene expression in the infected neuron and favor the establishment of latency [5] , [47] . As originally supposed by Roizman and Sears [48] , a lack of retrograde axonal transport of VP16 upon entry into the distal axons could be responsible for the poor onset of IE gene expression in neurons in vivo . By contrast , infection of the soma may enable VP16 to reach the nucleus and to initiate a lytic infection of the neuron , as observed in neuronal cultures non-selectively infected in vitro . In addition , transactivation by residual VP16 reaching the soma of the axonally infected neuron may be blocked . Modulation of the function of VP16 by other components of the HSV-1 outer tegument layer might contribute to the insufficiency of transactivation of IE gene expression in axonally-infected neurons . The tegument proteins UL46 and UL47 , which reportedly increase the activity of VP16 [49] , [50] , are predominantly lost from capsids during retrograde axonal transport [22] . In addition , residual VP16 activity may be blocked by the cytoplasmic sequestration of HCF-1 in unstimulated neurons [51] , and the interaction of HCF-1 with the basic leucine zipper proteins , Luman and Zhangfei [52]–[54] . Various cell culture systems have been used to study the establishment of and reactivation from latency at the molecular level in vitro [5] . If a missing or strongly reduced axonal transport of VP16 is important for the lytic/latent decision in the infected neuron , the ability of HSV-1 to induce lytic gene expression in neurons should differ with the site of viral entry . Specifically , the selective infection of distal axons should predispose neurons to a nonproductive HSV-1 infection . To our knowledge , and contrary to this assumption , differences specific to the site of entry in the ability of HSV to cause productive infection of neurons remain unknown . The selective entry of HSV-1 into the axons of dissociated DRGs in vitro has been reported to result in the productive infection of neurons , and consequently in the progressive spread of HSV throughout the cultures [42] , [55] . While a predominantly nonproductive HSV-1 infection can be spontaneously established in neuronal cultures [56] , acute lytic infection must usually be suppressed for some time to enable the efficient establishment of latent infection in vitro , especially if infections are performed at a high MOI [57]–[61] . As shown by Bertke et al . [62] neuronal factors may have a significant impact on the relative permissiveness for productive infection . Thus , adult murine A5-positive sensory neurons are nonpermissive for productive infection with HSV-1 in vitro . To our knowledge , chicken trigeminal neurons have not been tested for the expression of the A5 ( Galβ1-4GlcNAc-R ) epitope . As additional markers of adult murine trigeminal neurons preferentially harboring latent HSV-1 , the calcitonin-gene related peptide ( CGRP ) and the high affinity NGF receptor ( TrkA ) have been described [63] . The majority of neurons in the adult chicken trigeminal ganglion are reactive for the calcitonin-gene related peptide ( CGRP ) [64] . Furthermore , basic characteristics of neurotrophins and their receptors in sensory neurons appear to be well preserved between mammals and the chick , e . g . , the same genes are found and expressed with shared common features in different neuron populations during development . TrkA is expressed at high levels in the embryonic chicken TG [65] . Analysis of the viral transcript and genome levels in HSV-1 infected embryonic chicken TGEs between 6 hpi and 7 dpi revealed that the infection of free distal axons spontaneously resulted in a largely nonproductive HSV-1 infection that was characterized by stationary genome levels with a median of approximately 10 , 000 to 20 , 000 genomes/TGE infected with 5×106 pfu HSV-1 17syn+ , the expression of low and fluctuating levels of lytic transcripts , and the prominent expression of LAT at 24 hpi and thereafter . A general lack of susceptibility of the organ model to HSV-1 is unlikely to account for this observation . Previous studies showed that neuronal embryonic chicken tissues are permissive for HSV-1 , and are a suitable model system for the analysis of neurotropic alphaherpesviruses [44] , [66] . Our data confirm that HSV-1 can efficiently infect the distal axons of embryonic chicken TGEs . We estimate that the addition of HSV-1 at saturating infectious doses to the AC results in stable mean levels of 70–210 HSV-1 genomes per neuron with neurites reaching the AC . This is highly consistent with levels of approximately 70 and 180 viral genomes per latently infected LAT-negative and LAT-positive neuron in the mouse model reported by Chen et al . [67] . Fusion-incompetent viral particles added to the AC of embryonic chicken TGEs were not transported in a retrograde direction to the GC . Thus , in agreement with earlier studies , HSV-1 enters the distal axons of embryonic chicken TGEs by direct fusion with the axonal membrane , and not via endocytotic uptake [22] , [68]–[70] . If this is correct , a lack of axonal transport of VP16 should result in a similar pattern of lytic gene expression in neurons axonally-infected with wt HSV and a VP16AD-negative mutant . In fact , ICP27 transcript levels in CHX-treated TGEs axonally infected with HSV-1 KOS RP5 and RP5R did not differ . Furthermore , levels of lytic transcripts at 24 hpi were found to be comparable in cultures infected with HSV-1 17syn+ and various EGFP-expressing HSV-1 17 mutants , HSV-1 KOS RP5 and RP5R ( data not shown ) . Thus , at least in the early phase of infection , conspicuous VP16-dependent differences in lytic viral transcript levels were not evident in axonally-infected TGEs . Proenca et al . [71] recently demonstrated that HSV-1 IE gene expression by a VP16-independent mechanism can precede the establishment of latency in sensory neurons . In addition , we also found similar levels of LAT expressed by HSV-1 wt virus and the VP16AD-negative mutant in axonally-infected TGEs . In the absence of functional VP16 , HMBA has been described to enhance IE gene expression and lytic growth of HSV-1 , especially if cells are infected at low MOI [45] , [72] . The exact mode of action of the cytodifferentiating agent HMBA on HSV-1 growth is not fully understood . HMBA induces the differentiation of murine erythroleukemia cells ( MELC ) and other transformed cells to a less transformed phenotype [73] . Comparison of the effects of various other agents known to promote the differentiation of MELCs demonstrated that some of these substances complement the growth of VP16-deficient HSV-1 whereas others antagonize the effect of HMBA [74] . HMBA-treatment also increases the replication of VP16-positive HSV-1 and HSV-2 in epidermal and neuronal cells [75] , and the oncolytic activity of a gamma1-34 . 5-negative HSV-1 mutant in oral squamous carcinoma cells [76] . Accordingly , HMBA treatment significantly increased the expression of lytic genes in TGEs axonally-infected with the VP16AD-negative mutant HSV-1 KOS RP5 . HMBA treatment also restored the ability of VP16-competent HSV-1 strains to initiate a productive infection of sensory neurons and stimulated viral spread within nonneuronal cells . In TGEs axonally-infected with the spread-deficient mutant HSV-1 KOS gH87 , HMBA treatment significantly enhanced the expression of lytic genes ( data not shown ) . Together these findings provide evidence of insufficient transactivation of IE gene expression by VP16 in axonally-infected neurons . With respect to the ability of HSV-1 to induce a productive infection of neurons when added directly to the GC , we reasoned that silent HSV-1 infection resulting from axonal infection could be transactivated by the direct co-infection of neurons with HSV-1 helper virus . Our results show that both genome replication of the reporter virus and reporter gene expression under the control of HSV-1 IE , E/L , and human CMV ( HCMV ) IE promoters can be transactivated by simultaneous infection with a helper virus introduced into the GC . Quantification of helper and reporter virus levels in co-infected TGEs indicated that only a few helper virus particles appear to be sufficient to transactivate the reporter virus in neurons . These results further support the hypothesis that the poor expression of lytic genes after axonal entry is due to low or missing transactivation by VP16 , since helper-virus-induced transactivation of IE gene expression by simultaneous co-infection of the GC occurred in the absence of de novo protein synthesis , and efficient transactivation was dependent on the presence of functional VP16 in the HSV-1 helper virus . The related alphaherpesviruses HSV-2 and PrV also efficiently transactivated lytic HSV-1 gene expression in axonally-infected neurons . However , unlike HSV-1 and HSV-2 , the lytic gene expression after axonal entry of PrV is sufficient to initiate a synchronous , productive infection of neurons . The rapid spread of PrV in the explants following axonal infection is highly consistent with its high in vivo pathogenicity , which is substantially greater than that of HSV [77] . The simultaneous co-infection of the GC with HSV-2 or PrV was found to directly transactivate HSV-1 IE gene expression in neurons in the absence of de novo protein synthesis . Transactivation of HSV-1 IE gene expression is most likely mediated by HSV-2 VP16 and the PrV VP16/pUL48 homologue [78] , although it has been reported that UV-inactivated PrV particles do not transactivate the expression of reporter genes controlled by IE promoters of HSV-1 or PrV [79] , [80] . Interestingly , the HSV-2 and PrV but not HSV-1 helper viruses were still able to efficiently transactivate HSV reporter virus gene expression at 7 dpi . It is unclear why HSV-1 helper virus specifically fails to transactivate HSV-1 genomes in axonally-infected neurons when added after the establishment of a silent , latency-like infection in the cultures . However , our finding is in agreement with the results of Su et al . [81] , who reported that HSV-1 genomes in long-term NGF-differentiated , quiescently-infected PC12 cells become refractory to superinfection with HSV-1 . Mador et al . [82] showed that the expression of HSV-1 LAT renders neurons resistant to infection with HSV-1 but not HSV-2 , and concluded that protection against superinfection with HSV-1 may be an important function of HSV-1 LAT . Since axonally-infected TGEs express high levels of LAT at 7 dpi , a LAT-dependent mechanism may be responsible for the HSV-1-specific failure to transactivate silent HSV-1 infection observed in this study . As discussed by Berngruber et al . [83] , protection of the host cell harboring latent virus by superinfection inhibition is not restricted to bacteriophages but appears to be a common mechanism in the maintenance of latency in many viral systems including HSV-1 , and may be a driving force in the evolution of viral latency . The use of fluorescence microscopy to monitor intact , axonally-infected TGEs with the replication-competent mutant HSV-1 17 CMV-IEproEGFP showed that a few isolated neurons expressed EGFP within the first days after infection . Monitoring the onset of EGFP expression in individual neurons revealed that this was associated with an approximately fivefold increase in the cumulative number of EGFP-positive neurons during the first 10 days after axonal infection . The cumulative number of EGFP-positive neurons increased at a small but constant rate during weeks 3–10 after axonal infection . Balliet et al . [84] interpreted the de novo onset of EGFP expression under the control of the HCMV IE promoter in latently HSV-1-infected neurons as spontaneous molecular reactivation , a term coined by Feldman et al . [85] . Spontaneous molecular reactivation , as observed in the mouse TG , might reflect asymptomatic viral shedding and recurrent disease in humans occurring in the absence of known external stimuli of HSV reactivation [86] . Stochastic spontaneous reactivation patterns have also been described in other latent herpesvirus infections [87] . Since HSV-1 spreads from EGFP-positive neurons to surrounding nonneuronal cells in a small proportion of axonally-infected embryonic chicken TGEs , at least a fraction of the former must have been productively infected . Interestingly , after a short period of expansion , the growth of these plaque-like clusters of infected cells ceased and EGFP-positive cells finally disappeared again . This is reminiscent of the results of Barreca and O'Hare [88] , [89] , who reported that HSV-1-infected MDBK cells develop an interferon-dependent , refractory state leading to the suppression of viral growth on the one hand and the persistence of virus in a subpopulation of cells on the other . As reported by Camarena et al . [90] , continuous NGF-mediated signaling is required to maintain HSV-1 latency in primary neuron cultures . The use of serum-free culture media supplemented with NGF as the only growth factor may thus have favored the long-term nonproductive infection of neurons observed in the present study . The presence of fetal calf serum ( FCS ) in the culture medium and continuing viral spread from infected epithelial cells to distal axons instead may enhance the productive HSV-1 infection of embryonic chicken TGEs [43] . In the present study we found that treating the cultures with the cell-differentiation-inducing agent HMBA not only enhanced the expression of lytic genes in axonally-infected neurons but also strongly promoted secondary viral spread to the surrounding neurons and nonneuronal cells . Clearly , differentiation , trophic state , developmental stage , and age of neurons and nonneuronal cells significantly impact the outcome of HSV-1 infection in neuronal organotypic cultures . In particular , even if the vast majority of neurons become nonproductively infected , conditions favoring the massive secondary spread of HSV-1 will disrupt the establishment of a long-lasting latency-like HSV-1 infection as long as a few productively infected neurons and susceptible cells are present in the organotypic cultures . Thus , the apparent failure of HSV-1 to spontaneously establish a latency-like infection upon axonal infection in earlier studies may also reflect differences in the endogenous capability of cultures to establish a refractory antiviral state . In conclusion , our work provides direct experimental evidence that HSV-1 entry into the distal axons of sensory neurons is characterized by a site-of-entry-specific restriction of lytic gene expression . We consider our observations to be relevant for the course of natural HSV-1 infection and the underlying mechanism to be decisive for the establishment of latency . Although the block to lytic infection is not complete , switching off the transactivation of IE genes by VP16 is probably an essential first step in the combinatory circuit that ultimately triggers the establishment of latency in sensory neurons infected via the physiological route . In turn , the repression of massive lytic replication in the PNS despite infection of neurons at a high MOI may enable the host to initiate an innate immune response and prevent anterograde spread of the virus until adaptive immune mechanisms have developed [91]–[93] . In this way , both the efficient establishment of latent infection and the prevention of excessive viral pathogenicity might be warranted .
All procedures involving chicken embryos complied with the relevant national guidelines . Embryonated chicken eggs are not subject to the restrictions imposed by German laws related to animal protection . Furthermore , any organ explanted from a killed animal is not defined as an animal experiment ( Tierschutzgesetz §7 ( 1 ) ) . For infection experiments , explant cultures were treated with the virus , and thus no infection of live organisms occurred . The use of chicken embryos has been approved officially as an alternative to replace animal experiments , in accordance with the replacement , refinement , and reduction strategy for animal experiments of the EU . TGEs and standard cell lines were cultivated at 37°C in an atmosphere of 5% CO2 and 100% humidity . Vero and RK13 cells were grown in minimal essential medium ( MEM; Biochrom , Berlin , Germany ) supplemented with 10% v/v FCS and antibiotic/antimycotic additives ( 100 µg/ml streptomycin , 100 U/ml penicillin G , and 0 . 25 µg/ml amphotericin B; Invitrogen , Karlsruhe , Germany ) . TGEs were grown in TG medium consisting of d-valine-modified MEM ( Biochrom ) with 5 g/l glucose , 200 mM l-glutamine ( Sigma-Aldrich , Munich , Germany ) , 50 ng/ml rat NGF 7S ( Invitrogen ) , and antibiotic/antimycotic additives . The carboxymethylcellulose ( CMC ) -containing TG medium comprised MEM with 0 . 5% CMC ( Sigma-Aldrich ) , 5 g/l glucose ( Sigma-Aldrich ) , 50 ng/ml NGF 7S , 1% human serum ( Harlan Sera-Lab , Loughborough , UK ) , and antibiotic/antimycotic additives . The HSV-1 strain 17syn+ , EGFP-expressing HSV-1 recombinants and HSV-2 333 were propagated on Vero cells , and infection titers were quantified using a plaque assay . The gH-negative mutant HSV-1 KOS gH87 in which the 1 , 790-bp NcoI-XbaI-fragment of the gH gene has been replaced by the lacZ gene under control of the CMV IE promoter was kindly provided by P . G . Spear ( Chicago , IL , USA ) . Fusion-competent stocks of HSV-1 KOS gH87 were propagated and titrated on gH-expressing VeroF6gH cells kindly provided by A . C . Minson ( Cambridge , UK ) [97] . Fusion-deficient virus particles were collected from the supernatants of nontranscomplementing Vero cells . Enrichment and purification of HSV-1 particles from cell-culture supernatants by ultracentrifugation on a 20% saccharose cushion was performed as described previously [94] . The VP16AD-negative strain HSV-1 KOS RP5 and the corresponding revertant RP5R were generated and propagated as described by Tal-Singer et al . [46] , and kindly provided by S . J . Triezenberg , Van Andel Research Institute , Grand Rapids , MI , USA . For infection of TGEs , HSV-1 KOS RP5 and RP5R were grown and titrated on Vero cells . Using 204-nm latex polystyrene beads ( Plano , Wetzlar , Germany ) as a calibration standard , the particle/pfu ratios of virus stocks of HSV-1 KOS RP5 , HSV-1 KOS RP5R , and HSV-1 17syn+ propagated on Vero cells were determined to be 10 , 000 , 31 . 5 , and 18 , respectively . The particle/pfu ratios of purified particles of HSV-1 KOS gH87 propagated on VeroF6gH and Vero cells were 15 . 5 and 2×107 , respectively . Generation of the EGFP-expressing mutant PrV-KaΔgGgfp is described elsewhere [98] , and the virus was propagated on RK13 ( rabbit kidney ) cells . CHX , nocodazole , acyclovir , HMBA , and other chemicals used for cell culture were purchased from Sigma-Aldrich . Clinical isolates of HSV-1 and HSV-2 were obtained from the routine diagnostic laboratory of the University Hospital of Münster . Isolates were propagated on Vero cells as described above . Day-15 chick embryos were used for the preparation of TGs . The skull was opened dorsally , the brain removed , and the TG dissected out and maintained in MEM at 4°C until being used further . Tissue culture dishes ( 35 mm in diameter ) were coated with 10 mg/ml gelatin ( Biochrom ) followed by 20 µg/cm2 mouse laminin ( Roche , Mannheim , Germany ) . They were then washed and dried . The bottom of the coated tissue-culture dishes was scratched with a pin rake ( Tyler Research , Edmonton , Canada ) to form equally placed grooves . A cloning cylinder ( Omnilab , Bremen , Germany ) that was 8 mm high , and with an internal diameter of 8 mm and a wall thickness of 1 mm , was attached to the center of the culture dish with sterile silicone grease ( Bayer , Leverkusen , Germany ) , forming inner and outer compartments . TGEs were fixed to the inner chamber by 1–2 min of air-drying , and then both compartments were supplied with TG medium and incubated under standard conditions for 5 days . During TGE fixation , care was taken to ensure that the pole of explants containing axons projecting to the periphery in vivo faced the inner rim of the diffusion barrier . Axon penetration from the inner to the outer compartment was monitored visually prior to infection . For selective staining of neurons innervating the external compartment , 5 µl of the retrograde axonal tracer DiI ( Vybrant DiI cell-labeling solution , Invitrogen ) was added per milliliter of culture medium and incubated for 20 min . Before the selective infection of sensory nerve endings in the AC , the medium in the GC was replaced by 0 . 3 ml of CMC-containing TG medium ( as described above ) . After 30 min , the virus suspended in 2 ml of TG medium was added to the AC , where it was incubated for 1 h at 37°C . The virus inoculum was then removed , the AC washed once with phosphate-buffered saline ( PBS ) , and CMC-containing TG medium was added . Cross-contamination by leakage from the AC was avoided by keeping the fluid level slightly higher in the GC than in the AC . For nonselective infection of TGEs , the medium in the GC was carefully aspirated and virus suspension ( 0 . 3 ml ) was added . At 1 hpi , the virus inoculum was replaced by 0 . 3 ml of CMC-containing TG medium . For infection of the GC with helper virus , the medium of the inner chamber was carefully aspirated , and 0 . 3 ml of helper virus suspension was added . After 45 min , the helper virus inoculum was replaced by 0 . 3 ml of CMC-containing TG medium , in which cultures were incubated at 37°C . To avoid cross-contamination , co-infection of the AC with reporter virus was performed 20 min after completing infection of the GC with helper virus and the addition of CMC-containing TG medium to the GC . At 1 h after incubation of the AC with 2 ml of reporter virus inoculum at 37°C , the virus inoculum was aspirated from the AC , cultures were washed once with PBS , and CMC-containing TG medium was added . The chambers were incubated at 37°C for the duration of the experiments . TGEs were fixed at different times post infection for 3 h in 4% phosphate-buffered paraformaldehyde ( pH 7 . 4 ) at 4°C . Fixed TGEs were embedded in Tissue-Tek ( Sakura Finetek , Zoeterwoude , The Netherlands ) and frozen in isopentane cooled in liquid nitrogen . Cryosections were cut at a thickness of 7 µm on a cryotome ( CM3050 , Leica , Bensheim , Germany ) and mounted on coated slides ( Superfrost Plus , Langenbrinck , Emmendingen , Germany ) . Fluorescence immunohistochemistry was performed on frozen sections using previously described standard protocols [104] . The polyclonal anti-HSV-1 antibody ( Dako , Glostrup , Denmark ) was diluted to 1∶200 and incubated with the sections overnight at 4°C . A biotinylated α-rabbit secondary antibody ( Vector Laboratories , Burlingame , CA , USA ) was used , followed by fluorescence-signal detection with A488-streptavidin ( Molecular Probes , Leiden , Netherlands ) . Nuclei were stained using a mounting medium containing DAPI ( Vector Laboratories ) . Sections were examined with a fluorescence microscope ( DM , Leica ) , and images were digitized and transferred to a computer using a Diagnostic Instruments SPOT II camera system ( Visitron , Munich , Germany ) . The monochrome fluorescence signals were merged into a single multicolor image using SPOT II software . Preparations of dispersed TGEs were obtained , processed for immunofluorescence and stained with HSV-1-specific rabbit antiserum , a monoclonal antibody against the 200 kD neurofilament marker ( Roche ) and DAPI as described by Hafezi et al . [43] . Daily monitoring of EGFP expression was performed with a fluorescence microscope ( Axiovert 200 , Zeiss , Jena , Germany ) using a bandpass filter for EGFP ( AHF Analysentechnik , Tübingen , Germany ) . Images were obtained using an AxioCam MRm camera ( Zeiss ) and AxioVision 4 . 6 software ( Zeiss ) . | Upon primary infection of the oronasal mucosa , herpes simplex virus type 1 ( HSV-1 ) rapidly reaches the ganglia of the peripheral nervous system via axonal transport and establishes lifelong latency in surviving neurons . Central to the establishment of latency is the ability of HSV-1 to reliably switch from productive , lytic spread in epithelia to nonproductive , latent infection in sensory neurons . It is not fully understood what specifically disposes incoming particles of a highly cytopathogenic , fast-replicating alphaherpesvirus to nonproductive , latent infection in sensory neurons . The present study shows that selective entry of HSV-1 into the distal axons of trigeminal neurons strongly favors the establishment of a nonproductive , latent infection , whereas nonselective infection of neurons still enables HSV-1 to induce lytic gene expression . Our data support a model of latency establishment in which the site of entry is an important determinant of the lytic/latent decision in the infected neuron . Productive infection of the neuron ensues if particles enter the soma of the neuron directly . In contrast , previous retrograde axonal transport of incoming viral particles creates a distinct scenario that abrogates VP16-dependent transactivation of immediate-early gene expression and precludes the expression of lytic genes to an extent sufficient to prevent the initiation of massive productive infection of trigeminal neurons . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"animal",
"models",
"of",
"infection",
"viral",
"entry",
"host",
"cells",
"viral",
"persistence",
"and",
"latency",
"viral",
"transmission",
"and",
"infection",
"virology",
"neurovirulence",
"biology",
"microbiology"
] | 2012 | Entry of Herpes Simplex Virus Type 1 (HSV-1) into the Distal Axons of Trigeminal Neurons Favors the Onset of Nonproductive, Silent Infection |
The learning mechanism in the hippocampus has almost universally been assumed to be Hebbian in nature , where individual neurons in an engram join together with synaptic weight increases to support facilitated recall of memories later . However , it is also widely known that Hebbian learning mechanisms impose significant capacity constraints , and are generally less computationally powerful than learning mechanisms that take advantage of error signals . We show that the differential phase relationships of hippocampal subfields within the overall theta rhythm enable a powerful form of error-driven learning , which results in significantly greater capacity , as shown in computer simulations . In one phase of the theta cycle , the bidirectional connectivity between CA1 and entorhinal cortex can be trained in an error-driven fashion to learn to effectively encode the cortical inputs in a compact and sparse form over CA1 . In a subsequent portion of the theta cycle , the system attempts to recall an existing memory , via the pathway from entorhinal cortex to CA3 and CA1 . Finally the full theta cycle completes when a strong target encoding representation of the current input is imposed onto the CA1 via direct projections from entorhinal cortex . The difference between this target encoding and the attempted recall of the same representation on CA1 constitutes an error signal that can drive the learning of CA3 to CA1 synapses . This CA3 to CA1 pathway is critical for enabling full reinstatement of recalled hippocampal memories out in cortex . Taken together , these new learning dynamics enable a much more robust , high-capacity model of hippocampal learning than was available previously under the classical Hebbian model .
Over the past half century the hippocampus has provided fertile ground for the work of mechanistic computational models to inform empirical research . From the earliest investigations into Long Term Potentiation to the complex dynamics of place cells , models of hippocampal function have enabled a greater understanding of how learning and memory emerges from more basic neural mechanisms in this remarkable brain area . The paradigmatic theoretical model guiding this work is the Hebb-Marr framework [1]–[3] , which features the core idea that Hebbian learning wires together neurons that are firing together as part of a memory or engram representation , e . g . , in the central area CA3 of the hippocampus . With these connections strengthened , the ability to pattern complete a partial memory cue to a full representation of the original memory is enhanced . For this pattern completion within CA3 to actually drive full memory recall , it must trigger a chain reaction of pattern completion throughout the cortex — although central to most theoretical accounts , the critical role of the CA1 in this larger pattern completion process has not been as widely recognized . Specifically , learning between CA3 and CA1 neurons must take place at memory encoding , to enable the CA1 to then drive entorhinal cortex ( EC ) , which then drives the higher-level association cortex areas that are bidirectionally interconnected with it . This plasticity at the CA3 to CA1 synapses indeed may be the most important factor for subsequent memory recall [4] . It is the nature of this plasticity , and the learning that takes place in the bidirectional connections between EC and CA1 , that is the focus of this paper . We argue that , by taking into account the phase differences of firing for these areas within the overall theta cycle of the hippocampus [5] , [6] , a powerful error-driven form of learning emerges , which can result in much higher storage capacity than the standard Hebbian learning mechanism . Furthermore , these phase dynamics within the EC – CA1 bidirectional connections enable the CA1 to very naturally learn to be a sparse , invertible auto-encoder of the EC inputs , which has long been an important but somewhat implausibly implemented feature of our computational models [7]–[10] . Thus , this new model , which we refer to as the theta-phase hippocampus model , in reference to the theta oscillation , provides a more unified and computationally powerful model of hippocampal function . This model also enables us to make more direct contact with a large base of evidence , in both humans and rodents , relating hippocampal EEG oscillations to learning and memory . Much of the progress within this literature has been made in animal electrophysiology targeting hippocampal representation during spatial navigation and recall , while evidence from human EEG and intracranial recordings of oscillatory interactions also shows connections to episodic memory . Modeling work , originally developed within the spatial navigation literature , suggested that connectivity between hippocampal subregions is coordinated via the 3 to 8 Hz EEG theta oscillation [5] , [6] . This work has also been extended into a more general framework of hippocampal function including a proposed extension from spatial navigation into episodic memory [8] , [11] . These investigations provide the foundation for the theta-phase model described in the current work , in terms of establishing the existence and functional role of the oscillatory coordination of hippocampal subregions within an encoding and retrieval dynamic . We build upon this foundation by showing how these dynamics can lead to error-driven learning , and a concomitant increase in overall storage capacity for the system . The implementation of this theta-phase model is based directly on the Complimentary Learning Systems neural network model of the hippocampus [7] , [9] , [10] , which is implemented within the Local , Error-driven , and Associative , Biologically Realistic Algorithm ( Leabra ) framework [12] , [13] . We assess the impact of the theta-phase error-driven learning mechanisms by comparing it with an otherwise identical model that uses a Hebbian learning rule , while varying the number of units within the Dentate Gyrus ( DG ) and area CA3 , and measuring the models' recall on a varying number of learned patterns . These learned patterns are presented at test with 25 percent of the pattern missing , and the models are compared on their ability to complete the missing portion of the pattern . Results show the error-driven signal performs significantly better than the Hebbian learning rule .
The model used in the current work is built upon a series of structural and functional hypotheses based on anatomical and physiological data , which have been captured in the complementary learning systems ( CLS ) model of the hippocampus [7] , [9] . The Entorhinal Cortex ( EC ) in the model is assumed to be the cortical gateway to the hippocampus . This gateway feeds through the trisynaptic pathway ( TSP ) to the Dentate Gyrus ( DG ) , CA3 , and then to CA1 . Similarly , there is a parallel connection through the monosynaptic pathway ( MSP ) from the EC to the CA1 ( and back ) ( Figure 1 ) . The TSP connections via the perforant path from EC to DG and CA3 are broadly diffuse , and support the conjunctive binding of various distributed pieces of information into an overall episodic memory representation in the CA3 . The CA3 has sparse and highly separable patterns of activity ( which are further pattern-separated via the very sparse DG layer ) , resulting in substantially reduced interference from synaptic weight changes , thus enabling rapid learning of novel episodic or conjunctive information [7] . To recall existing memories , the recurrent connections in CA3 , along with plasticity in the EC to CA3 , as well as the DG to CA3 connections , support pattern completion of missing information from retrieval cues . For pattern completion in CA3 to have any effect on the rest of the brain , there must be a way to map the CA3 representation back out to the neocortex . This occurs via connections from CA3 to CA1 ( the Schaffer collateral pathway ) , and then from CA1 back to EC , which then projects back out to the cortex to fill in the full memory representation in the cortical areas where it can actually be used in further cognitive processing . This Schaffer collateral pathway is a key focus of the theta-phase model , where we can train synapses in this pathway according to an error-driven learning signal , instead of the standard Hebbian signal assumed in other existing models . The MSP between EC and CA1 is also essential for supporting memory retrieval , in a way that is often under-appreciated in the literature . This pathway is topologically organized , not diffuse , which we capture by organizing the simulated neurons in EC and CA1 into mutually interconnected slots , presumably encoding different separable elements across all the cortical areas that converge on the EC [14] . This slot architecture ( Figure 1 ) enables the MSP to develop separable invertible pathways where a given EC input pattern can be encoded over a sparser representation in the corresponding CA1 slot , and this CA1 representation can in turn recover the full original EC slot pattern . The topographic nature of this CA1 representation is important for providing a mapping from cortex into the hippocampus and back out again . Weight adjustments along the TSP form conjunctive representations that bind information across the topography of EC and are important for recreating a previously experienced state from incomplete inputs ( i . e . , pattern completion ) . The Schaffer collaterals ( the connection between CA3 and CA1 ) provide the translation between these two types of representations , allowing the conjunctive representations learned in the TSP to influence the topographic representations within CA1 , and subsequently back out to EC . In our previous CLS models , we have trained these topographic slot mapping weights between EC and CA1 in an offline manner prior to training the full hippocampal network . The new theta-phase learning mechanisms now enables us to train this important MSP pathway in a very natural manner , at the same time as the rest of the hippocampal system learns . To summarize , after learning , the model recollects studied items by reactivating the original patterns via the trained weighted connections between areas . The accuracy of this recall is scored as a simple comparison between the originally studied pattern and the recollected pattern . If the input pattern corresponds to a non-studied pattern , or even if individual components of the pattern were previously studied , but not together , the conjunctive nature of the CA3 representations will minimize the extent to which recall occurs . Conversely , when previously studied patterns are presented in an incomplete or noisy input format , these weights allow the hippocampus to recall the originally studied pattern . As noted previously , the original Complementary Learning Systems ( CLS ) hippocampal model pretrained the invertible mapping between EC and CA1 on a vocabulary of possible patterns for a single slot [9] . The resulting weights for the connections within this individual slot network were then replicated across all EC–CA1 slots ( see Figure 1 where an individual slot is highlighted ) in the MSP . This restricts the space of inputs possible to the vocabulary of patterns in which the slot network was trained . The alternative approach adopted in this work utilizes simultaneous , independent learning along both the MSP and the TSP . This dual-pathway learning is motivated by physiological recordings within the subfields of rat hippocampi , along with mathematical models of hippocampal function [6] , in terms of the 3–8 Hz oscillatory EEG signal known as theta . The theta oscillation can be found throughout the hippocampus and surrounding cortex , however it is strongest and most consistent when recorded within the region separating CA1 and DG known as the hippocampal fissure . For this reason all references to theta oscillations will be referring to the EEG signal measured at the hippocampal fissure . Figure 2A shows an illustration of hippocampal subfield dynamics in relation to the fissure recorded theta oscillation shown in red . This cartoon , derived from current source density analysis [6] , [15] , shows the current sinks into area CA1 alternatively originating from either area CA3 in blue or EC layers II and III in green . At the trough of fissure recorded theta , EC sources into CA1 are at their peak and area CA3 is at its minimum . This implies that EC has a strong influence over synaptic potentials within area CA1 at this time . At the peak of fissure recorded theta , CA3 sources are at their peak and EC influence has diminished . This again suggests that CA3 input to area CA1 is now the dominant influence , and EC is less so as compared to the trough of the theta oscillation . These dynamics are modeled within the neural network as , for any given input pattern , three distinct time points of activation: Theta Trough ( TT ) , Theta Peak ( TP ) , and Theta Plus ( + ) as shown in Figure 2B . These three time points are modeled as three independent settling processes across simulated neurons within the differential equation described in eq . ( 1 ) . The patterns of activation that arise from these three time points are used to train the weighted connections along the MSP and the TSP , where the equations for the error-driven weight changes at these synapses are shown in eqs . ( 6 ) and ( 7 ) . Specifically , input patterns are projected onto which is then allowed to project to CA1 and subsequently to area , while CA3 input to CA1 is inhibited . This creates a pattern of activation dominated by the MSP which is then used to drive learning within these connections . This is denoted as superscript TT in eq . ( 6 ) for “Theta Trough” , as this time point is analogous to the connectivity dynamics at the trough of theta oscillations , where EC strongly influences CA1 , and CA3 influence is relatively low . Following this , CA3 input onto CA1 is released from inhibition while the influence from onto CA1 is diminished . This corresponds to the Theta Peak ( denoted as TP in eq . ( 7 ) ) ; a time point that reflects strong influence from CA3 onto CA1 . This time point is analogous to the peak of the fissure recorded theta oscillation where EC input to CA1 is weak , while CA3 input is strong . The final plus stage of activation ( denoted with the + in eqs . ( 6 ) and ( 7 ) ) corresponds to projecting onto and area CA1 , and projecting back onto CA1 . The representations within and will remain relatively static due to the direct connection between them , which then forces CA1 to settle into a representation that respects this symmetric mapping between and . This provides the veridical ground truth in the error-driven learning signal . In reference to eq . ( 3 ) , this pattern of activation is used for the plus stage learning signal in contrast to the MSP's TT and the TSP's TP minus stage . The alteration of these connections' strength are manipulated in the model by simply denying information flow through specific subregion projections at select points in the settling process of the differential equation shown in eq . ( 1 ) . The three particular projections that are manipulated in the model are , and where the pattern of manipulation that these projections are subjected to are highlighted in Table 1 . All other connections within the network have no error-driven component to their weight adjustments , only Hebbian , as seen in eq . ( 8 ) . The validation process adopted in this work is to compare the theta-phase learning model described above with a simple Hebbian learning model . The critical connections that utilize an error-driven learning signal within the theta-phase model are the Mono-Synaptic Pathway ( ) , as well as the Schaffer collaterals ( ) . In contrast , these connections in the comparison model use a purely Hebbian learning rule . The task run across both models is a simple capacity test such that each model is trained for 15 repetitions of an input pattern set ( referred to as 15 epochs ) , and the performance of the two models is then tested by measuring the accuracy of the recalled patterns of activation given an input cue which has 25 percent of the trained pattern missing . We explored three training regimes to contrast error-driven vs . Hebbian learning . First , both the MSP and TSP utilized an error-driven learning signal and was compared to a full Hebbian network . We then compared the contribution of these two pathways by using error-driven learning within either the MSP and not the TSP , or conversely within the TSP and not the MSP . Finally , to better compare against earlier models where the MSP pathway was pretrained in advance , we compared pretrained vs . non-pretrained MSP . In the pretrained MSP , only the MSP pathway was trained for 15 epochs ( on the same patterns used for the overall training ) , followed by integrated training of both TSP and MSP as described above . In the non-pretrained MSP , both pathways were trained in the integrated fashion from the start . The question of how network performance scales is addressed by varying the training set size , and network size across multiple levels of these two variables . The size of the input pattern set is varied from 40 to 800 patterns to get a measure of model performance across small and large training sets , with the assumption that better performance on larger training sets is more reflective of hippocampal function . Similarly , the size of the network itself was varied by increasing the number of units within the CA3 and DG layers , while holding a constant ratio between them . This is done to try and maintain a connection to the original biological constraints of the hippocampal circuit , and for this reason a ratio of 5 DG units to 1 CA3 unit was adopted , as this generally reflects the ratio in the human hippocampus [14] , [16] . Maintaining this ratio , the total units within CA3 were varied from 10 to 100 units , which in turn corresponds to a varying of DG units from 50 to 500 . Finally , input patterns were constructed , and memory retrieval performance measured , based on the slot topology in the EC layers ( as highlighted in Figure 1 ) . This slot structure is intended to capture the modality segregation within EC , and within each slot we assume there is a vocabulary of different patterns , which reflect the representational repertoire within those modalities . We generated a vocabulary of 100 distributed activity patterns , with a minimum hamming distance of 10 between each vocabulary pattern generated . A complete input pattern used in the model validation process was then constructed by selecting a single pattern from these 100 vocabulary patterns for each of the EC slots . With 8 slots , a total of 1008 ( npatterns raised to the nslots power ) unique patterns are possible , however only 800 were used in the testing of these models . These vocabulary patterns were similarly used to estimate error within the networks' output by comparing , within a given slot , the output pattern of activation with all other vocabulary patterns . If , for the given input pattern , the slots' output at the layer is closest to the vocabulary pattern it was trained on , it is considered correct , and otherwise considered incorrect . This closest-pattern calculation is done across each of the slots for every input pattern , and if any slot shows an incorrect response the network output for that input pattern is counted as incorrect . This measurement is referred to as Name Error in the results section , and is thought to better represent the potential for clean up of hippocampal output as compared to more standard measures such as Sum Squared Error ( SSE ) . It also has the advantage of not requiring any further threshold or other parameterization . It should be noted that this measure of error , compared to a SSE , deemphasizes single unit based errors in output in favor of an emphasis on distributed patterns of error across groups of units . The model is implemented in the Leabra framework which uses a combination of supervised and Hebbian learning [13] . What follows is a coarse description of the essential components within this framework necessary for understanding the current work . The activation function for a given unit is a threshold based neuronal model with continuous valued spike rate as output . Each neuron's membrane potential ( ) is updated using the following differential equation: ( 1 ) Here , 3 channels ( ) summed across in the membrane potential calculation are: excitatory input , inhibitory input , and leak current . Excitatory input is calculated as the average over all weighted inputs coming into a unit ( ) , where is the activity of sending unit and is the weighted connection between sending unit and receiving unit . All principal weights between units are excitatory while local circuit inhibition controls positive feedback loops . Leabra assumes a winner take all dynamic through a set-point inhibitory current ( ) , producing a kWTA ( k-Winners-Take-All ) dynamic . kWTA is computed via a uniform level of inhibitory current for all units within a layer . Finally leak current ( ) is a constant value set to 0 . 1 Activation of communication ( for a given unit ) with other units is a thresholded function of membrane potential: ( 2 ) Here is the gain factor which is set to a constant value of 100 , and is the firing threshold value which is set to a constant of 0 . 5 within a units dynamic range of 0 to 1 . The Leabra framework utilizes a biologically plausible error-driven learning algorithm which is equivalent to Contrastive Hebbian Learning ( CHL ) [17] . Leabra uses two stages of activation; the stage is the initial activation or expected output of the network , while the stage is the provided target output activation . The Leabra weight updating component between sending units ( ) and receiving units ( ) is thus calculated as: ( 3 ) The + and − superscripts represent the plus and minus phase components respectively . In addition to the error-driven learning of CHL , a pure form of Hebbian learning is also used . Here the weight change is calculated using only the target , or plus phase , activations ( 4 ) This learning rule can be seen as computing the expected value of the sending unit's activity conditional on the receiver's activity [13] . Finally these two learning rules are proportionally weighted ( ) along with a learning rate parameter , , for the combined learning rule used in this work: ( 5 ) The theta-phase learning approach uses the learning framework described above within a particular dual-pathway architecture . The target , or plus , component of the error signal ( superscript ‘+’ in eqs . ( 6 ) and ( 7 ) ) is activation acquired from the layer projected onto , and allowed to propagate back on to area CA1 which settles into a pattern of activation constrained by static representations in and . Similarly projects along the TSP providing a plus phase activation within DG and CA3 , however projections from CA3 onto CA1 are inhibited . Error signals used in weight adjustment are then calculated by taking the difference between this plus phase activation and the two distinct time points within the theta cycle ( peak and trough ) , yielding two distinct error signals . Specifically , the MSP connections are adjusted according to an error signal acquired from the difference between plus phase activation and activation patterns acquired during the trough of theta ( superscript TT in eq . ( 6 ) ) . It is critical to remember in the trough of theta there is no influence on representations from area , while in the plus phase projects onto . The difference in CA1 activation patterns at the conclusions of these two phases allows for the calculation of an error signal that is used to adjust the weighted connections within the MSP . Similarly , the TSP connections are adjusted according to an error signal acquired from the difference between the plus phase activation and the activation during the peak of theta ( superscript TP in eq . ( 7 ) ) . In the peak of theta CA3 has a strong influence on CA1 , while in the plus phase CA1 is influenced solely by the MSP . This change in CA1 representations allows for a error signal tailored to best adjust the TSP connections to more closely match the stimulus driven representation of the plus phase activations . All other connections , within the network , i . e . to DG and CA3 , DG to CA3 , and recurrent connections within CA3 , have no error-driven component to their weight adjustment ( eq . ( 8 ) ) . ( 6 ) ( 7 ) ( 8 ) Settling dynamics within the network are dictated by the temporal evolution of Equation 1 . This dynamic process , within every unit , is allowed 30 time steps to settle into its equilibrium state for each of the three phases within the theta cycle , thus yielding a total of 90 time steps for each full theta oscillation . All activation values within the network are reset to 0 at the onset of theta trough but are allowed to be carried over from trough to peak and finally from peak to the plus phase without alteration . All manipulations of Hebbian vs . error-driven learning where done via the lmix parameter as shown in eq . ( 5 ) . Values used to instantiate full Hebbian learning implies a lmix value of 1 , while error-driven learning used a lmix value 0f 0 . 001 . This implies that error-driven networks also used a very small amount of Hebbian weight adjustment which we believe is implicit in normal neural circuitry .
Figure 3 shows the comparison of various network configurations . In panel A the theta-phase network with error-driven learning in the MSP and TSP is compared with a fully Hebbian learning network across various network sizes and trained input pattern set sizes . Plots are shown as a function of network size , where the number of CA3 units are shown on the x-axis which implies that the number of DG units for that network are 5 times that of CA3 . Training set size , shown on the y-axis refers to the number of patterns a given network was trained and tested on . Surface plots of the average Name Error , on the z-axis , across the full training set are shown on the left for both the theta-phase and the Hebbian network . Each cyan dot in the surface plots represents a measured data point where both network types were tested in the network-size by training-set-size space . Each data point is the average within network type across 5 random weight initializations . These points were then fit to a 3D surface for visualization . The difference between theta-phase and Hebbian surfaces is shown on the right . These differences are compared using a random bootstrap method where Name Error values are sampled with replacement from both network types into two groups and a distribution of difference values is calculated to produce a null hypothesis . Data points with p values less than 0 . 005 are shown in the difference plot with an asterisk . Similarly , in panel B of Figure 3 a more fine grained follow up test of performance shows a network with error-driven learning in the TSP and Hebbian learning in the MSP ( labeled TSP ErrDrv ) compared against a network with Hebbian learning in the TSP and error-driven learning in the MSP . This secondary test attempts to evaluate the relative importance of error-driven learning within these two pathways on overall performance . These results are then further tested by comparing pretrained MSP connections to non-pretrained MSP connections . These results are not shown in a similar style as Figures 3A and B as these pretrained networks yielded results nearly identical to non-pretrained networks . Figure 3C shows performance plots from individual networks within these three comparisons; here the overlap in pretrained and non-pretrained results can be seen in comparison to the other two comparisons shown in Figures 3A and B . Results , shown in panel A of Figure 3 , of the comparison between network models shows that the error-driven learning , provided by the theta coordination of subfield influence on CA1 , out-performs the purely Hebbian based learning network . Investigating this relationship further , in panel B of Figure 3 it is shown that the crucial connection that leads to this benefit is between CA3 and CA1 along the TSP . There is little difference in performance when the TSP uses Hebbian learning ( plot labeled MSP ErrDrv ) compared to when the full network is exclusively using Hebbian learning . Conversely , when the TSP ( specifically the connection between CA3 and CA1 ) takes advantage of the error-driven learning signal , performance is dramatically increased , and approaches Name Error levels achieved when both TSP and MSP are using error-driven learning signals ( shown in Figure 3B and C ) . Contrasting performance from the full ThetaPhase network with the TSP error-driven network shows that there is indeed some performance benefit in the ThetaPhase network compared to TSP error-driven network , suggesting some synergy between the TSP and MSP over and above the benefit from the TSP error-driven network alone . Our comparison of the effects of pretraining on the MSP , as was done in our earlier models , revealed very little difference as shown in Figure 3C . This is of considerable practical benefit , as it is often difficult to anticipate the full range of input pattern variability needed for pretraining , and it also increases the overall plausibility of our model , by eliminating any need for this extra step in the model . These results provide insight as to how these learning signals compare across multiple network sizes and varying training set sizes . Looking at the difference in network performance we can see a divergence towards better error-driven performance as training size and network size increases . Many hippocampal models used within the literature test on relatively small training sets and with small network sizes; usually of the size required for the task or phenomena being modeled . Results from the current work suggest that Hebbian performance may not scale with these dimensions as expected , and that a more robust learning signal such as that provided by error-driven learning may be necessary to provide realistic performance in more ecologically valid network sizes and training set sizes . Given the significant performance advantages of the error-driven learning mechanism , and its biological support in the theta-phase coordination process , it would be surprising if the biological hippocampus did not also leverage this form of error-driven learning . In sum , we argue that this model represents a significant advantage over the existing Hebbian-based models of hippocampal learning , and can provide a predictive framework for future empirical studies . The idea of temporal differentiation between Mono-Synaptic and the Tri-Synaptic pathways along the theta wave , as shown in previous hippocampal modeling work [5] , [6] , provides a well founded framework for how theta oscillations interact with behavior . The key contribution of this work to these models is a demonstration that the invertible mapping in and out of area CA1 along the Mono-Synaptic Pathway can be learned in tandem with the connections along the Tri-Synaptic Pathway , and that these oscillatory dynamics enable a form of powerful error-driven learning . Further , these results suggest that error-driven learning in the Schaffer collaterals connecting CA3 to CA1 are a crucial component in stabilizing this invertible mapping in the Mono-Synaptic Pathway , and providing the performance advantage shown in Figure 3 . The mapping of distributed representations into and out of area CA1 is a problem that has not been adequately addressed in previous models . Many models have used a simplified symmetric representation between hippocampal subregions [5] , [6] . This allows for a transparent interpretation of subregion processing , however it reduces the ecological validity of the model's processing . An early model of episodic memory allowed for learning within this invertible mapping between EC and CA1 , however the representations used were relatively small and simplified [8] . The current work shows that error-driven learning is a key component behind the requirement of relatively complex representational transformation between subregions . The attempt to match the hippocampal architecture and representational complexity within this work provides insight into these more subtle issues that are often assumed in other models of the hippocampus . The simulations done in this work show that the representational transformation into and out of the hippocampus is a non-negligible problem , and that more robust learning signals than the standard Hebbian model are required for accurate recall within large training data and small network sizes . The current model provides a simplified version of oscillatory processes within a discretized time frame , as compared to previous models [5] , [6] . The peak and trough time points being modeled in the current work can be thought of as stimulus driven at the trough of theta , and recall driven at the peak of theta [6] , however these processes are implemented within the model as two relatively discontinuous patterns of activation that get integrated together when calculating the weight changes in the learning algorithm . Additionally , the plus phase of activation , i . e . the ground truth within the error calculation , is proposed as a projection of the superficial EC layers onto the deep layers of the EC . Computationally within the model this is implemented after both the trough and peak of the theta oscillation have completed , however we conceptualize the theta cycle to begin on the trough of the oscillation where the MSP is strongly active , and we therefore speculate that this plus phase projection would occur within the descending theta cycle following the peak but just before the trough . In Figure 2 we show the plus phase to occur at the trough of theta , however the model predicts that the plus phase would occur anytime between theta peak and theta trough . In some sense the plus phase is a transition from theta peak to theta trough where the onset of the plus phase is marked by the inhibition of the TSP and a projection from the superficial layers of EC to the deep layers . This allows for the error-driven contrasting of this plus phase pattern of activation with the preceding theta trough and theta peak patterns . Indeed , laminar recordings from Entorhinal Cortex support this theta phase reversal in deep layers compared to superficial [18] , and a recent investigation into the microcircuits within EC layers supports the increased firing from superficial EC to deep EC just preceding the trough of ongoing theta oscillations [19] . Future electrophysiological work could test these temporal dynamics further by stimulating at these various stages of the theta wave to try and disrupt or enhance this theoretical cascade of activation . Previous models have labeled activation patterns associated with theta peak and trough as Encoding during the trough , and Retrieval during the peak , which our model also captures [5] , [6] , [20] . This separation of functionality between the two pathways might allow for other systems to interact with the nominal theta cycle to influence these processes and thereby bias the hippocampus towards one process over the other . A growing base of empirical evidence within the rodent literature suggests that oscillatory coherence within the theta band between frontal regions and the hippocampus is correlated with successful retrieval [21]–[23] . In humans these interactions could provide the framework for some form of volitional control over either encoding or retrieval . Future empirical work in humans could probe this relationship between encoding and retrieval within the hippocampus as well as its interaction with other systems . The current model would suggest that disruption of the theta oscillation during the trough of theta would alter the encoding of new experiences , while disruption at the peak of theta would alter the retrieval of previous experiences . The question of how incoming stimuli align to these phase dynamics is somewhat unclear , however constraints from previous empirical work do exist . There is evidence suggesting that theta oscillations show a phase resetting approximately 200 ms after stimulus onset [24]–[26] . The entry point into the theta wave on these phase resets , however , show a difference in study vs . test items where test items enter on the descending wave of theta while study items enter on the ascending wave . Our model suggests that there would be a plus phase following the descending theta wave , and would be evident through the projection from the superficial layers of EC to the deep layers . This task dependent phase reset could help to target this plus phase dynamic , and potentially determine whether it is more associated with start of a given theta oscillation or with the end . There are many limitations within the current work in regards to the scope of biological components , and we do not mean to suggest that this model accurately reflects all aspects of hippocampal function . For example , the discrete nature of the two time points modeled , i . e . trough and peak , within the theta cycle could be better approximated by having a continuous change of activation after the plus phase . The current work simplifies the more continuous change of activation at the end of a Theta cycle by resetting activation after the plus phase . Additionally , there are hippocampal subfields , in particular the Subiculum [27] , which are not included within this model . We are currently exploring the addition of a Subiculum layer within our model which modulates the learning rate of connections into CA1 . The Subiculum-mediated modulation focuses on increasing the learning rate for novel stimuli , and reducing the learning rate for well learned Tri-Synaptic Pathway ( TSP ) representations , theoretically allowing for the reduction of interference in the otherwise purely Hebbian learning in the TSP ( e . g . , in perforant pathway projections from EC to CA3 ) . Although no current explorations are underway , area CA2 could also provide an augmentation to our model of the MSP [28] . This area would fit in as a intermediary between the and CA1 , providing a non-topographic representation across the slots of Entorhinal Cortex , and potentially increasing the learning capacity along this pathway . In conclusion , within the subfields modeled , we have accurately represented the known connectivity and topology using a biologically motivated neural network framework . Further , we have included coordination between those subfields through the currently understood inhibitory processes as modulated by theta oscillations . Building upon this framework in future projects can provide a strong foundation in the known biological constraints , and representational complexity of the hippocampal circuit . | We present a novel hippocampal model based on the oscillatory dynamics of the theta rhythm , which enables the network to learn much more efficiently than the Hebbian form of learning that is widely assumed in most models . Specifically , two pathways , Tri-Synaptic and Mono-Synaptic , alternate in strength during theta oscillations to provide an alternation of encoding vs . recall bias in area CA1 . The difference between these two states and the unaltered cortical input representation creates an error signal , which can drive powerful error-driven learning in both Tri-Synaptic and Mono-Synaptic pathways . Furthermore , the presence of these alternating modes of network behavior ( encoding vs . recall ) provide an intriguing target for future work examining how prefrontal control mechanisms can manipulate the behavior of the hippocampus . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results/Discussion"
] | [
"circuit",
"models",
"computer",
"science",
"psychology",
"computer",
"modeling",
"social",
"and",
"behavioral",
"sciences",
"cognitive",
"psychology",
"neural",
"networks",
"computational",
"neuroscience",
"memory",
"biology",
"computerized",
"simulations",
"neuroscience",
"learning",
"and",
"memory",
"coding",
"mechanisms"
] | 2013 | Theta Coordinated Error-Driven Learning in the Hippocampus |
Strongyloides stercoralis is a widely distributed parasite that infects 30 to 100 million people worldwide . In the United States strongyloidiasis is recognized as an important infection in immigrants and refugees . Public health and commercial reference laboratories need a simple and reliable method for diagnosis of strongyloidiasis to identify and treat cases and to prevent transmission . The recognized laboratory test of choice for diagnosis of strongyloidiasis is detection of disease specific antibodies , most commonly using a crude parasite extract for detection of IgG antibodies . Recently , a luciferase tagged recombinant protein of S . stercoralis , Ss-NIE-1 , has been used in a luciferase immunoprecipitation system ( LIPS ) to detect IgG and IgG4 specific antibodies . To promote wider adoption of immunoassays for strongyloidiasis , we used the Ss-NIE-1 recombinant antigen without the luciferase tag and developed ELISA and fluorescent bead ( Luminex ) assays to detect S . stercoralis specific IgG4 . We evaluated the assays using well-characterized sera from persons with or without presumed strongyloidiasis . The sensitivity and specificity of Ss-NIE-1 IgG4 ELISA were 95% and 93% , respectively . For the IgG4 Luminex assay , the sensitivity and specificity were 93% and 95% , respectively . Specific IgG4 antibody decreased after treatment in a manner that was similar to the decrease of specific IgG measured in the crude IgG ELISA . The sensitivities of the Ss-NIE-1 IgG4 ELISA and Luminex assays were comparable to the crude IgG ELISA but with improved specificities . However , the Ss-NIE-1 based assays are not dependent on native parasite materials and can be performed using widely available laboratory equipment . In conclusion , these newly developed Ss-NIE-1 based immunoassays can be readily adopted by public health and commercial reference laboratories for routine screening and clinical diagnosis of S . stercoralis infection in refugees and immigrants in the United States .
Strongyloides stercoralis , an intestinal nematode that migrates through the skin and lung , is a widely distributed disease that infects 30 to 100 million people worldwide [1] . Unlike other helminthic parasites , S . stercoralis can complete its entire life cycle within a single human host through autoinfection and can cause an asymptomatic chronic infection that may go undetected for decades in immunocompetent hosts [2 , 3] . In the United States , S . stercoralis causes more deaths than any other soil-transmitted helminth , with mortality rates as high as 87% in cases of hyper-infection in immunocompromised hosts [3] . The standard diagnosis of strongyloidiasis relies on the detection of larvae in the stool [4] , but a single stool sample analysis will identify no more than 70% of positive cases [5] . Due to the low sensitivity of the stool assay , immunodiagnosis using a crude antigen-based enzyme-linked immunosorbent assay ( ELISA ) has been developed as the laboratory test of choice for clinical diagnosis of strongyloidiasis . The Immunoglobulin G ( IgG ) ELISA utilizes crude extract prepared from L3 S . stercoralis larvae obtained from infected dogs . Reliance on native parasite materials and the canine infection model are major disadvantages of this test . As a result , a number of recombinant antigen-based ELISAs have recently been developed . Recombinant antigens can be purified easily and can be reproducibly generated in large amounts [6–8] . Antibody detection assays utilizing recombinant protein Ss-NIE-1 , a 31-kDa antigen derived from S . stercoralis L3 parasites [8] , have reported sensitivities and specificities of 84–98% and 95–100% , respectively , and are comparable in performance to the crude antigen-based ELISA [6–13] . We have incorporated Ss- NIE-1 into a standard ELISA format assay and into a fluorescent bead format assay ( Luminex ) to detect S . stercoralis-specific Immunoglobulin subtype G4 ( IgG4 ) . We have previously used the Luminex system for the simultaneous determination of IgG antibody responses to multiple infections in a single assay run [14–17] and we hope to add the new multiplex bead antibody test to our Neglected Tropical Disease assay panel . We compared the performance of the Ss-NIE-1 recombinant antigen-based ELISA and Luminex bead assays to the published assay performance parameters for the Ss-NIE-1 luciferase immunoprecipitation system ( LIPS ) based assay [6 , 10] and the crude antigen-based IgG ELISA [10] . Because previous research has documented that not all cases of strongyloidiasis are successfully treated with a single course of therapy [18] , we also used the fluorescent bead assay to determine if a decrease in antibody was measureable after treatment using a select set of sera .
Although some samples were exhausted during the initial ELISA development and some new samples were added during Luminex assay development , the same sets of sera were used for testing the Ss-NIE-1 ELISA and Ss-NIE-1 Luminex assays and many samples were assayed using both techniques . The sets of human sera used were: ( 1 ) samples proven positive for S . stercoralis based on the presence of larvae in the stool or sputum ( ELISA N = 258 , Luminex N = 175 ) ; ( 2 ) presumed negative samples from U . S . residents with no history of foreign travel ( ELISA N = 182 , Luminex N = 207 ) ; ( 3 ) a convenience panel of samples from patients with various diseases other than S . stercoralis focusing mainly on worm infections and including 63 sera from proven cases of lymphatic filariasis from Haiti ( ELISA N = 143 , Luminex N = 159 ) [19]; ( 4 ) and sera from patients with S . stercoralis infections , before and after treatment ( ELISA N = 48 , Luminex N = 25 ) [18] . All sera were anonymous and were used in accordance with approved human subjects’ protocols . The development of an IgG4 standard reference curve for the Ss-NIE-1 ELISA was performed as described by Scheel et al [22] . Immunoglobulin G4 , human myeloma plasma was purchased and stored frozen in 20 mM phosphate , pH 7 . 4 , with 150 mM NaCl and 0 . 05% Sodium azide ( NaN3 ) ( Athens Research & Technology , Athens , GA ) . IgG4 was diluted into antigen sensitizing buffer ( ASB ) ( 0 . 05 M Tris/HCl , pH 8 . 0 + 1 M KCl + 2 mM EDTA ) to create standard curve points . IgG4 concentrations were chosen based on previous experiments in standard curve development and adjusted to produce the highest OD value , ~ 2 . 0 . Checkerboard titrations for antigen concentration , serum dilution , conjugate dilution , and substrate 3 , 3’ , 5 , 5’-Tetramethylbenzidine ( TMB ) time were carried out on Immulon 2HB Microwell plate ( Thermo Scientific , Cat . Number 6506 ) . For optimization of the Ss-NIE-1 ELISA , optimal conditions were chosen based on the signal to noise ratio between defined strong S . stercoralis positive and normal human serum samples . The optimized Ss-NIE-1 ELISA steps are as follows: the micro-well plate was sensitized with 100 μL/well of Ss-NIE-1 antigen at a concentration of 0 . 3 μg/mL in antigen sensitizing buffer ( 0 . 05 M Tris/HCl , pH 8 . 0 + 1 M KCl + 2 mM EDTA ) for 2 hours at room temperature on a plate shaker . Following antigen sensitization , the plate was washed 4 times with PBS/0 . 3% Tween . The plate was blocked for 30 minutes with 100 μL/well of 10 mM Nickel Chloride ( Aldrich , Cat . Number 339350 ) in PBS/0 . 3% Tween/5% Instant Nonfat Dry Milk ( Nestle , Glendale , CA ) , and then washed as before . StabilCoat Immunoassay Stabilizer ( SurModics , East Prairie , MN ) was then added 100 μL to each well and incubated for 30 minutes on a plate shaker at room temperature . After discarding the blocking solution , the plate was dried for 4 hours at 30°C in a vacuum oven chamber . The sensitized and blocked plate was stored at 4°C in sealed aluminum foil with desiccator . Human serum samples were tested in 100 μL/well at 1:50 dilution in PBS/0 . 3% Tween/5% Instant Nonfat Dry Milk . Following 30 minutes incubation at room temperature on a plate shaker ( speed ~ 800 rpm ) , the plate was washed 4 times with PBS/0 . 3% Tween . Proper conjugate concentration of mouse anti-human IgG4 ( clone HP6025 ) , affinity purified , horseradish peroxidase labeled ( Southern Biotech , Birmingham , AL; Cat . Number 9200–05 ) was added to each well at 100 μL/well at 1:1 , 000 dilution in PBS/0 . 3% Tween and incubated 30 minutes at room temperature on a plate shaker with the plate being washed 4 times following incubation with PBS/0 . 3% Tween . The substrate used was SureBlue , 3 , 3’ , 5 , 5’-Tetramethylbenzidine ( TMB ) Microwell Peroxidase Substrate ( KPL , Gaithersburg , Maryland ) . We used 100 μL/well of TMB to develop the plate for 5 minutes , and the reaction was stopped by adding 100 μL 1 N H2SO4 Analyzed ACS Reagent ( J . T . Baker , Phillipsburg , NJ ) . The signal was read at A450nm using a VersaMax Kinetic ELISA Microplate Reader with SoftMax Pro v5 . 4 Software ( Molecular Devices Corporation , Palo Alto , CA ) . Data were tabulated and analyzed using Microsoft Excel . Determination of the cut-off value and assay performance was calculated using SAS version 9 . 0 . The concentration of Ss-NIE-1 ELISA was measured in in ng/mL; Luminex results are reported as mean fluorescence intensity minus background blank ( MFI ) . The J-index , a single measurement of assay performance , was calculated as described previously ( [25 , 26] .
The recombinant Ss-NIE-1-His protein expressed at Genscript was successfully used to develop an ELISA ( Table 1 ) . Using a cutoff value of 0 . 80 ng IgG4/mL , the assay correctly identified 245 of 258 parasitologically confirmed strongyloidiasis cases for a sensitivity of 95% ( Table 1 ) . The overall specificity of the ELISA was determined to be 93% using a panel of non-endemic US controls and sera from patients with other ( mainly parasitic worm ) diseases ( Tables 1 and 2 ) . Among the 142 donors with defined diseases or parasitic infections shown in Table 2 , the specificity was only 86% , but only the two trichuriasis patients were 100% cross-reactive . Although the numbers of samples were also quite small , cross reactivities of ≥ 33% were observed among sera from echinococcosis , gnathostomiasis , and hookworm patients . Table 2 shows the cross-reactivity of the various presumed negative sera . Given the likelihood of polyparasitism among many of these serum donors ( i . e . , 63 lymphatic filariasis patients from S . stercoralis-endemic Haiti ) , the 99% specificity observed among US negative controls may be a reasonable upper bound to the value . The J-index , a single measure of assay performance , was 0 . 88 . Positive sera with low and medium level reactivity were used to measure inter-assay variation as previously described in the Materials and Methods . The inter-assay coefficient of variation was determined to be 22% for the low positive control serum and 10% for the medium positive control serum . The His tagged Ss-NIE-1 recombinant protein could not be coupled to magnetic beads . Thus , we proceeded using a GST-tagged Ss-NIE-1 protein and successfully coupled the Ss-NIE-1 protein to the MagPlex microspheres . The intra- and inter-assay coefficients of variation were determined to be 4 . 2% and 13 . 9% , respectively , for average values of 641 MFI and 585 . Using a cutoff value of 8 MFI , the sensitivity and specificity of the IgG4 Luminex assay were 93% and 95% ( Table 1 ) , respectively . As with the ELISA described above , the specificity among US negative controls was much higher ( 99% ) than among donors with defined diseases or parasitic infections ( 91% ) ( Table 2 ) . High reactivities ( ≥33% ) were only observed among sera from the two amebiasis and three hymenolepiasis donors . The J-index was identical to that of the ELISA at 0 . 88 . Antibody longevity in subjects infected with strongyloidiasis following treatment with thiabendazole [18] can be seen in Table 3 . Peak and median antibody responses decreased over time using both assay formats , but the antibody levels remained above the cut-off point for most of the subjects even 18 months post treatment . Using the Kobayashi criteria of cure , which considers a patient to be cured if the ratio of serological results post-treatment compared to pre-treatment is less than 0 . 6 , 70% of the subjects would be reported as cured 3–6 months following treatment [27] .
Strongyloidiasis is an increasingly important health problem in the US among immigrants and refugees . Patients with occult strongyloidiasis are at risk of disease if they become immunosuppressed , and organ donors with unrecognized S . stercoralis infection pose a risk to recipients if their infected organs are transplanted [2] . Identification of the parasite in stool specimens is insensitive , and , because parasitological examination requires collection of multiple stool specimens over 3 days , serological testing would be preferred if available . We elected to develop novel immunodiagnostic assays to meet this need . We employed a well described recombinant protein with proven performance as an immunodiagnostic antigen , the Ss-NIE-1 protein [6–8] . Based on the potential importance of IgG4 antibody responses in filarial infections , we decided to develop methods that detect S . stercoralis-specific IgG4 antibodies . The Ss-NIE-1 IgG4 Luminex bead assay achieved a sensitivity and a specificity comparable to those reported for other strongyloidiasis assays such as the crude antigen ELISA , and 26-kDA ELISA , and Ss-NIE-1 ELISAs and the Ss-NIE-1 LIPS [6 , 7 , 9 , 10 , 12 , 18] . Compared to the CDC S . stercoralis crude antigen ELISA , the Ss-NIE-1 IgG4 ELISA and the Ss-NIE-1 IgG4 Luminex assay achieved similar sensitivity without compromising specificity . The possible factors contributing to improved specificity could be the use of recombinant antigen , assay optimization , or detection of IgG4 versus IgG . During Ss-NIE-1 ELISA optimization , we found that non-specific antibody binding in the normal human sera could be decreased by adding a pre-blocking step with 10 mM nickel chloride in PBS/0 . 3% Tween/5% milk . The decrease in background noise allows the assay to have a higher specificity ( 93% ) . ELISA based tests can only be used to detect antibody responses against one antigen of interest at a time . Because the differential diagnosis of S . stercoralis often includes multiple helminths , a submitted serum sample must frequently be tested using several parasite-specific ELISAs to determine the possible cause ( s ) of infection . xMAP Luminex technology offers an assay platform that can simultaneously detect antibodies to multiple diseases/infections , and multiplex bead-based antibody assays are generally as sensitive as conventional ELISA , have a wide dynamic range , and are highly reproducible from assay to assay [28] . For these reasons , we elected to transfer the Ss-NIE-1 assay to the Luminex platform . In our hands , the Ss-NIE-1 IgG4 Luminex bead assay was slightly less sensitive and slightly more specific than the ELISA , but had comparable overall performance as measured by the J-index . With the exception of toxocariasis and lymphatic filariasis , our cross-reactivity data must be interpreted with some caution due to the small numbers of samples available for testing . When sera from echinococcosis , hookworm , and trichuriasis patients were tested with Ss-NIE-1 by ELISA , 33% or more of the sera reacted , some quite strongly . A portion of this reactivity likely resulted from ELISA-specific background as many of these same sera did not react with the Ss-NIE-1 antigen in our newly developed Luminex bead assay . Bisoffi et al . [9] found no ELISA or LIPS assay cross-reactions to Ss-NIE-1 among their Echinococcus- or hookworm-infected donors . Although true cross-reactivity between these parasites may exist , polyparasitism cannot be ruled out in the remaining Luminex-positive subjects . An analysis of Ss-NIE-1 LIPS results from a hookworm/ Ascaris lumbricoides/ H . nana/ S . stercoralis co-endemic region of Argentina failed to demonstrate an association between infection with other parasites and an antibody response to the Ss-NIE-1 antigen [10] . The Ss-NIE-1 ELISA and Luminex had 10% cross-reactivity against the subjects with lymphatic filariasis . Again , this problem could be a true cross-reactivity or could also be explained by polyparasitism with soil-transmitted helminthes in Haiti . Unfortunately , we have no data about the presence of other parasitic infections in the individuals from whom these serum samples were obtained [29] . Norsyahida [11] reported that although the use of IgG4 conjugate did decrease the cross-reactivity to filariasis compared to the total IgG responses , some cross-reactivity was found with the IgG4 based assay . However , the Norsyahida group did not use a recombinant antigen [8] , and the observed cross-reactivity could be due to the use of crude extract antigen in their assay . Of note , the group mentioned the importance of testing for filariasis in subjects with strongyloidiasis . Such testing could most easily be accomplished on a multiplexing platform such as Luminex that can detect antibodies against filariasis and strongyloidiasis simultaneously . Decreases in antibody titers post-treatment were observed ( as suggested by Satoh [30] ) and the percentage of patients who met the serologic definition of cure ( ≥40% decline in antibody response compared to pre-treatment ) were consistent with the previous reports [18] . The only significant difference was that the CDC S . stercoralis crude ELISA results showed that 92%of subjects at >9–18 months had been cured of S . stercoralis infection , but only 71% could be considered as cured based on the SS-NIE-1 ELISA and Luminex assay . We have no definitive explanation for these observed differences , except the fact that the crude ELISA uses a complex antigen versus a single antigen which was used in our studies . Overall , excellent assays for detecting S . stercoralis specific antibodies have been developed . Because these assays use recombinant proteins , negating the need for native parasite materials these assays can be adopted for use in public health laboratories for refugee screening or in commercial laboratories for diagnosis of clinical strongyloidiasis and for screening possible transplant donors with occult disease . As both ELISA and Luminex based assays performed similarly , studies in low infrastructure , endemic setting could use the ELISA format . Although polyparasitism is a potential problem , with a strong specificity of 93% , cross-reactivity should not be an issue . For a country-wide study to determine the prevalence of strongyloidiasis , a multiplexing capability of Luminex will be more cost efficient . | Strongyloidiasis is a neglected tropical disease that affects millions worldwide and needs more attention and better diagnostic methods . Strongyloides stercoralis can undergo an autoinfection cycle and can cause hyperinfection involving the pulmonary and gastrointestinal systems and disseminated infection in other organs . Although endemic areas are mostly developing countries in tropical and subtropical regions with only sporadic transmission in temperate areas , the disease is a threat to developed world populations through immigrants , refugees , travelers , and military personnel . The disease can have catastrophic effects when a patient is immunocompromised or when an infected organ is transplanted into a vulnerable recipient . Due to the threat to public health , the intricate life cycle of S . stercoralis , the need to perform multiple follow-up diagnostics to ensure treatment success , and the necessity to rule out multiple co-endemic parasitic infections , it is imperative to develop new diagnostic assays that are simple and efficient while retaining maximal sensitivity and specificity . In this study , we use a well-known recombinant protein , Ss-NIE-1 , to optimize assays using both an ELISA format and a multiplex platform to meet these needs . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Development of Ss-NIE-1 Recombinant Antigen Based Assays for Immunodiagnosis of Strongyloidiasis |
The diagnosis of Chagas disease is complex due to the dynamics of parasitemia in the clinical phases of the disease . The molecular tests have been considered promissory because they detect the parasite in all clinical phases . Trypanosoma cruzi presents significant genetic variability and is classified into six Discrete Typing Units TcI-TcVI ( DTUs ) with the emergence of foreseen genotypes within TcI as TcIDom and TcI Sylvatic . The objective of this study was to determine the operating characteristics of molecular tests ( conventional and Real Time PCR ) for the detection of T . cruzi DNA , parasitic loads and DTUs in a large cohort of Colombian patients from acute and chronic phases . Samples were obtained from 708 patients in all clinical phases . Standard diagnosis ( direct and serological tests ) and molecular tests ( conventional PCR and quantitative PCR ) targeting the nuclear satellite DNA region . The genotyping was performed by PCR using the intergenic region of the mini-exon gene , the 24Sa , 18S and A10 regions . The operating capabilities showed that performance of qPCR was higher compared to cPCR . Likewise , the performance of qPCR was significantly higher in acute phase compared with chronic phase . The median parasitic loads detected were 4 . 69 and 1 . 33 parasite equivalents/mL for acute and chronic phases . The main DTU identified was TcI ( 74 . 2% ) . TcIDom genotype was significantly more frequent in chronic phase compared to acute phase ( 82 . 1% vs 16 . 6% ) . The median parasitic load for TcIDom was significantly higher compared with TcI Sylvatic in chronic phase ( 2 . 58 vs . 0 . 75 parasite equivalents/ml ) . The molecular tests are a precise tool to complement the standard diagnosis of Chagas disease , specifically in acute phase showing high discriminative power . However , it is necessary to improve the sensitivity of molecular tests in chronic phase . The frequency and parasitemia of TcIDom genotype in chronic patients highlight its possible relationship to the chronicity of the disease .
Chagas disease is a zoonotic parasitic disease caused by the protozoan Trypanosoma cruzi . It is considered a public health problem in Latin-America , where approximately 6 million people are currently infected [1] . The acute phase of the disease is characterised by usually mild fever that in a small proportion of cases can be accompanied by myocarditis and other lethal complications . Most of the patients continue through the chronic phase that is initially characterised by an asymptomatic clinical course during two or three decades , and about 30% of the infected patients will develop heart or digestive complications afterwards [2] . T . cruzi parasite shows significant genetic variability and classified into at least six Discrete Typing Units TcI-TcVI ( DTUs ) , that present associations with the geographical distribution , epidemiological transmission cycles , insect vectors and clinical manifestations of Chagas disease [3–5] . Recent studies suggest the occurrence of an emerging clade within TcI named TcIDom which is distributed in the Americas and associated with domestic cycles of transmission and human infection [6–10] . Recently , a genotype detected in anthropogenic bats and named as TcBat has been described in Panama , Ecuador , Colombia and Brazil including a case of human infection in Colombia [11–14] . The diagnosis of Chagas disease is complex due to the dynamics of parasitemia in the phases of the disease . During the acute phase the parasitemia is high , therefore the diagnosis is performed by direct parasitological tests [15 , 16] . Nevertheless , direct parasitological tests are not useful in the chronic phase due to the low and intermittent parasitemias . Therefore , the diagnosis of Chagas disease in the chronic phase is determined by serological tests such as ELISA: enzyme-linked immunosorbent assay , IFA: indirect immunofluorescence assay or HAI: Hemagglutination Inhibition Test [17–19] . Recently , molecular techniques such as cPCR ( conventional PCR ) and qPCR ( quantitative real-time PCR ) have been considered as supportive diagnostic tests due to their ability to determine parasitic loads of T . cruzi in all clinical phases of the disease [20–22] . The operating characteristics of molecular tests for diagnosis of T . cruzi infection have varied according to clinical phase and technical specifications . Sensitivity for identifying chronic infection with cPCR has ranged between 22 and 75% [23 , 24] and in both cases with a specificity of 100% . Contrastingly , for qPCR , sensitivity has ranged between 60 and 80% [22 , 25 , 26] in chronic phase and between 88% and 100% for acute phase [25 , 26] , whereas specificity is between 70–100% [26–28] . Sampling methods have not been always clearly stated and the role of these techniques for diagnosis of Chagas disease in the different clinical phases still remains poorly understood . The objective of this work was to determine the operating capabilities of qPCR and cPCR targeting the satellite nuclear DNA region , compared with standard diagnosis methods for acute and chronic Chagas disease . Additionally , we evaluated the plausible associations between parasitic load and DTUs in Colombian patients from the acute and chronic phases to untangle the natural course of T . cruzi infection in terms of parasite dynamics .
All patients who attended the Colombian National Health Institute ( Overall 985 individuals ) seeking diagnostic tests for Chagas disease in acute ( 113 patients ) or chronic phase ( 872 patients ) between 2004 and 2015 were considered as potential participants . Inclusion criteria were: i . Clinical or epidemiological suspect of Chagas disease in acute or chronic phase ii . Not having received aetiological treatment for Chagas disease iii . Positive serological tests for Chagas disease ( IFA , ELISA and/or HAI ) iv . Adequate blood and serum samples available for performing diagnostic tests according to the clinical phase . v . Acceptance to participate and sign the informed consent . The Technical Research Committee and Ethics Research Board at the National Health Institute in Bogotá , Colombia approved the study protocol CTIN-014-11 . Participation was voluntary and patients were asked for informed written consent authorising to take blood and serum samples and access information on their clinical records . The total sample size ( N ) was calculated for test binary outcomes and separately for each clinical phase: acute and chronic . Considering , n = Z2 S ( 1−S ) d2 , where for a confidence level of 95% ( 1- α , with α = 0 . 05 ) Z is inserted by 1 . 96 , and a maximum marginal error of estimate , d , is a desired value for precision based on researchers judgment , and S is a pre-determined value of sensitivity [29] . Based in previous studies , for the acute phase S was pre-established at 92% and with d at 8% [25 , 26] , whereas for chronic phase S was pre-established at 60% with d at 5% [22–26] . Then , N = n /P , being P the estimated prevalence in this specific population under study . Given this is a selected population , composed of patients with some suspicion of the infection and remitted to a reference centre , P was specified at 60% in suspected cases for both acute and chronic phases . This value was obtained as an approximation based on the laboratory records at the NHI ( Bogota , Colombia ) . The minimum total sample sizes were then calculated as N = 74 and N = 615 for suspected cases in acute and chronic phases respectively . The tests were performed to all subjects without knowing their previous clinical status . Clinical evaluation was conducted simultaneously to all individuals as part of the study to determine health status and then to the confirmed cases to evaluate heart complications . The inclusion of participants was conducted retrospectively for the period 2004 to 2012 , and prospectively between 2013 and 2015 . At the end , a total of 86 suspected acute patients and 622 suspected chronic patients were included in the study ( Table 1 ) . Operating characteristics of the molecular tests were estimated by comparing against standard diagnosis ( described above ) . Sensitivity , specificity , positive ( +LR ) and negative likelihood ratio ( LR- ) , predictive values ( PV ) , diagnostic precision ( DP ) , Area under the curve ( AUC ) , and Kappa index ( K ) were estimated for each phase of the disease ( acute and chronic ) , the clinical stage of chronic patients ( determined and undetermined ) and according to DTUs and TcI genotypes identified ( TcI sylvatic/TcIDom ) ( S2 Appendix ) . Results are presented as percentages , with corresponding 95% confidence intervals ( 95% CI ) . Additionally , operational capabilities in chronic patients were calculated in two ways: the first including negative patients without risk factors since they are the true negative and the second including all the negative patients ( with and without risk factors ) . Due to over dispersion of parasitic loads , medians and quartiles are presented . Comparisons are based on Mann-Whitney test between clinical phases , chronic clinical stages and the different T . cruzi DTUs and genotype groups identified . A p value at <0 . 05 was considered as statistically significant . All analysis was performed in Stata: Data Analysis and Statistical Software version 12 .
Overall , 985 patients were included , 872 suspected of chronic and 113 of acute infection . General demographic characteristics are shown in Table 1 . Out of the initial potential participants , 27 and 129 were excluded for incomplete samples to perform all analysis from the acute and chronic groups , respectively and 121 from the chronic group due to absence of clinical information ( Fig 1 ) . The inclusion of patients was prospective , whereas the sample collection was both retrospective ( for the period 2004–2011 ) and prospective ( for the period 2012–2015 ) . This means that for the retrospective component the samples were part of the repository . The repository consists of 144 samples , collected between 2004 and 2011 , and corresponds to serum samples stored at ( -80°C ) . In these samples , serological tests were repeated and it was found that the results were the same that they had been reported at the time of collection of samples and molecular tests were performed . The prospective component consists of 564 samples , collected in the period between 2012 and 2015 , and maintained in guanidine hydrochloride solution until processing . In patients from the acute phase , the qPCR test was positive in 95 . 7% of the patients and cPCR in 84 . 5% . In patients from the undetermined chronic phase , qPCR was positive in 68 . 0% of the cases and in 55 . 4% by cPCR . In the cardiac chronic phase , qPCR positivity was 59 . 1% and 58 . 6% by cPCR . The positive samples for satellite nuclear PCR ( qPCR and cPCR ) , were confirmed by kPCR . In patients that were negative by serology but with risk factors cPCR ( 2 . 6% ) and qPCR ( 3 . 6% ) were positive . In febrile and negative patients without risk factors both tests were negative in all samples . In all samples analyzed we detected the internal amplification control for both cPCR and qPCR , the average Ct value in all samples tested was 21 . The operating characteristics including all negatives patients of chronic phase ( Negatives with and without risk factors ) are presented in Tables 2 and 3 . Performance of qPCR was higher compared to cPCR in both acute ( AUC 0 . 98 vs 0 . 92 ) and chronic phase including only negatives with risk factors ( 0 . 82 vs 0 . 78 ) ( Fig 2 ) . Likewise , the performance was significantly higher in acute compared with chronic phase and in overall a specificity higher than sensitivity particularly in chronic phase ( Tables 2 , 4 and 5 ) . Parasitic loads were determined in samples that tested positive by qPCR . Significantly different median values were detected in acute ( 4 . 69 parasite equivalents/mL ) versus chronic phase ( 1 . 33 parasite equivalents/mL ) . A statistically median difference was also found between determined and undetermined chronic phase ( Fig 3 ) . In samples that tested positive ( n = 407 ) by cPCR , the DTUs TcI-TcVI and TcI ( TcI Dom , TcI Sylvatic ) were evaluated . The distribution of DTUs was 74 . 2% for TcI , 17 . 2% for TcII , 1 . 48% for TcIII , 0 . 5% for TcV and 6 . 7% for mixed infections . For the latter seven different combinations were identified: TcIDom/TcII/TcV , TcIDom/TcII , TcIDom/TcISylv , TcIDom/TcISylv/TcII , TcIDom/TcISylv/TcIII , TcIDom/TcIV , TcISylv/TcII . With respect to TcI , the genotyping was feasible in 290/302 samples . Out of them , 28 . 7% were classified as TcI Sylvatic and 71 . 4% as TcIDom . The median load parasitic value for TcII ( 4 . 68 parasite equivalents/mL ) was significantly different to the one for TcI ( 2 . 87 parasite equivalents/mL ) and TcIII ( 1 . 72 parasite equivalents/mL ) ( Fig 4 ) . The genotype distribution according to clinical phase evidenced that TcIDom was significantly more frequent in chronic phase compared with acute phase ( Table 6 ) . The operating characteristics of molecular tests for the different genotypes were calculated , observing that the sensitivity for identifying TcII was slightly higher than for TcI , mainly for qPCR ( S1 Table ) . The median parasitic load for TcIDom was significantly higher ( 2 . 58 parasite equivalents/ml ) compared with TcI Sylvatic ( 0 . 76 parasite equivalents/ml ) in chronic phase ( Fig 5 ) .
The main limitation involved in this study is the fact that there is not a gold standard test for all clinical phases of Chagas disease . Particularly for chronic phase , the best comparators are serological tests but these techniques measure the immune response and not the relative presence of the parasite . This particular situation impacts the evaluation of new diagnostic tests . This is reflected mainly in the kappa index ( Tables 2 and 4 ) that presented very low values in the undetermined and determined chronic phases . Unfortunately , it has not a simple solution and more understanding of the course of the infection is still needed . The results obtained for the molecular diagnosis in acute phase were optimal in terms of sensitivity for both qPCR ( 95 . 7%; 95%CI: 88 . 3–98 . 5 ) and cPCR sensitivity ( 84 . 5%; 95%CI: 74 . 3–91 . 2 ) , and same specificity . Although the results are showing a potential superior performance of the sensitivity of qPCR compared with cPCR , this difference needs a cautious interpretation . This might be explained due to the fact that detection by qPCR increases the sensitivity and specificity because of the hybridization of the Taqman probe in the amplicon , whereas in the case of the cPCR it requires a considerable amount of amplicon so that it can be observed in agarose gels [25 , 26] In addition , the confidence intervals were slightly overlapped , meaning that there is some indication of this difference but it is not statistically significant , so not definitive . The performance of the molecular tests in the acute phase is explained because there are large numbers of parasites , for example in cases of reactivation in immunosuppressed patients and in oral outbreaks . The values obtained for LR evinced the high probability that positive results correspond to diseased patients ( LR+ ) and the low probability that the diseased patients present negative results ( LR- ) . In addition , the DP was very optimal specifically for qPCR test confirming that this molecular test is very useful for the diagnosis in the acute phase , considering that the direct diagnosis is complex when the parasitemia is low ( As is the case of the acute patients detected more than a month after the infection where the parasitemia normally begins to decrease due to the control of the immune response ) and are required many tests for the confirmation of the acute cases ( direct tests , serology tests and clinical information ) . Regarding the predictive power of molecular tests in the acute phase , these tests are very good predictors of the disease presence when positive results are obtained ( PPV ) but their performance as predictors of absence of the disease are less ( NPV ) . However , it is worth noting that the predictive values depend on disease prevalence in the evaluated population . The analysis of operational capabilities in the chronic phase was conducted in the first instance including only negative patients without risk factors or true negatives . For the chronic phase , qPCR sensitivity was 64 . 2% and 56 . 8% for cPCR and in concordance with previous reports obtained by qPCR that have shown sensitivity ranging from 60–80% and 20–70% for cPCR [22–24 , 26 , 28 , 42] . These sensitivity results may be due to low and intermittent parasitic loads during chronic phase . The performance of qPCR was better than cPCR in the chronic undetermined phase , while that was very similar between the two tests in the determined chronic phase ( Tables 3 and 5 ) . The discriminative power of the two molecular tests was acceptable in the chronic phase . For qPCR , the AUC and DP values obtained ( Tables 3 and 5 ) were better for the undetermined phase than for determined phase . The differences between undetermined and determined phases for qPCR of the chronic phase can be explained by the natural course of the disease , in which the parasitic load decreases while increases the infection time . This is supported by several studies showing that there is no relationship between the evolution of the cardiac form of the disease and parasitemia but it declines with time as observed in this study [43 , 44] . Also , some studies show that cardiac form is mainly related to different types of strains , increased parasitemia , reinfection or immune system disorders in chronic patients [45 , 46] . In the cPCR AUC values were the same for both phases , while the value of DP was best for the determined phase . Possibly , this is because the detection limit of the cPCR is lower than qPCR , for this reason the cPCR behaves similarly in the two phases . In the two stages of the chronic phase , there is a high probability that patients with negative results in the molecular tests have the disease ( LR- ) and these tests are not good predictors of the absence of the disease ( NPV ) ( Table 5 ) . Therefore , the use of molecular methods as diagnostic tests is not appropriate due to the better performance displayed by serology . The probability that the results are positive is high in diseased individuals with respect to healthy individuals ( LR + ) and the molecular tests are excellent predictors of the presence of disease ( PPV ) . Thus , these tests could be used in situations in which the diagnosis is doubtful , allowing the confirmation of the parasite in diseased patients , which is of great importance for example when monitoring etiological treatment . However , it is necessary to improve the sensitivity , which can be performed by analysing serial samples for each patient as seen in some studies in which such sensitivity improved from 69 . 2% to 85 . 2% with the addition of a second sample or conducting DNA extraction from a larger volume of the sample [47 , 48] . In addition , the operating capabilities of patients in chronic phase were calculated including all negatives by serology with and without risk factors ( Table 1 , N = 141 ) . It was observed in the group of negative patients with risk factors a positivity of 2 . 6% ( 3 patients ) by cPCR and 3 . 6% ( 4 patients ) by qPCR , possibly due to an immunosuppression issue in these patients preventing the detection of antibodies or infection . Three patients are from the department of Casanare , which is an endemic area , and five patients had less than 24 years of age suggesting a recent infection . Also , all patients reported to know the vectors and have lived during his/her childhood in homes with features such as thatched or ‘barheque’ , floor or wood and/or tread walls of earth , wood or ‘barheque’ . Two of the seven patients that were negative by serology and had risk factors , whose ages were 36 and 51 showed the presence of symptoms at cardiac level . In this group of 7 patients , 4 presented the ELISA absorbance values greater than 0 . 200 and 4 detectable titles in the IFA ( 1/8 and 1/16 ) . As the operating capabilities calculated including all negative patients , a small percentage of decreased specificity in the two platforms was observed ( S3 Appendix ) . The positivity of these serologically negative patients that generated the decrease can probably be explained because cases of recent infection or patients with some form of immunosuppression that has generated the absence of detectable antibodies . In fact , in the group of acute patients , 4 patients whose serology was negative showed positive PCR , in these patients the detection was achieved by direct parasitological methods . Regarding the molecular techniques , given that in all PCR runs were included negative controls including reagents controls , a plausible contamination with parasite DNA is discarded . Significantly , the DP and AUC values showed no obvious changes unlike the values obtained for the NPV and the Kappa index , in which there was a marked increase . However , the changes obtained do not change the interpretation of the usefulness of the test in the clinical setting , but can show that there are few cases where serological tests may have false negatives as noted previously using cPCR by Ramirez et al . , 2009 [23] . Even though serological tests are considered the best current option for the diagnosis of Chagas disease , in a meta-analysis of high quality tests their sensitivity has been estimated at 90% [49] . Given this , we believe that an improvement of diagnostic tests for Chagas disease is needed for both serology and PCR techniques . An appropriate use of the comparator as gold standard and the inclusion of different phases of the disease are crucial to understand the utility of different diagnostic tests . To our knowledge , this is the first study to include statistical calculation of the sample , which allowed the analysis of operating characteristics of the molecular tests in all clinical phases of Chagas disease . In addition , this study is the first in analysing the two PCR platforms ( qPCR and PCR ) for the same target ( stDNA ) in patients from all clinical phases of Chagas disease . The conventional technique was included , due to the vast use of this technique in the diagnosis and its ease implementation in laboratories with restricted equipment ( a Real Time PCR machine is not available ) [23 , 24 , 28] . Lastly , acute patients had a less median age than chronic phase patients and in turn the largest number of acute cases are male . This possibly is because economic activity in endemic areas is developed by males that assist to the field and this facilitates direct patient contact with the vector and therefore with the parasite . On the other hand , females ratio and the median age were higher in chronic phase patients that are usually detected by screening blood banks or present cardiac abnormalities in chronic phase , then the detection occurs at a greater age . Additionally , in Colombia most blood donors are women facilitating their diagnosis . Regarding the parasitemia , it is observed that the median parasitemia was higher in acute patients compared to chronic phase , which is expected given the dynamics of parasitemia in the disease [25 , 26] . As for the group of chronic patients , the herein reported median of parasitemia is similar to those previously reported for Colombia [22 , 26] . In addition , the difference in medians between cardiac chronic and undetermined chronic stages was statistically significant , being higher in the undetermined chronic phase unlike the findings described by Ramirez et al , 2015 [26] , in which statistically significant difference was not detected . However , our results are in accordance with the natural history of the disease where parasitic loads decrease with the chronicity of the infection and this is probably associated with the type of strain and/or the immune response [2] . The DTU with highest frequency was TcI , both in acute and chronic patients , consistent with findings previously reported in Colombia [8 , 39 , 50 , 51] . Followed by TcII most often detected in chronic than acute patients . These findings are congruent due to the predominance of TcII in domestic cycles of transmission for the case of Colombia [50] . Regarding the parasitic loads of the DTUs detected , we observed that TcII had higher median parasitemia than other DTUs , consistent with the number of copies that has been reported in the DNA nuclear satellite region being higher for TcII than for TcI [52–54] . These findings highlight the importance of using the most representative DTU to generate the standard curves for quantification [22 , 25 , 26] . In addition , in murine models TcII shows higher parasitemias than TcI when performing individual and mixed infections [55] . In this study , acute cases are likely caused by vector transmission and possible oral route . In most of the cases TcI ( TcI sylvatic ) , TcII and TcIII infection was observed . These findings are consistent with previously documented reports for acute patients where DTUs associated with the sylvatic cycle of transmission were depicted [4 , 5 , 40 , 51 , 56–61] . An interesting finding was the detection of TcV in the patients surveyed . This DTU has been already reported in dogs and Rhodnius prolixus from eastern Colombia but this would be the first report of TcV human infection in the country [50] . It is necessary to conduct further studies to understand the host-parasite associations of this foreseen DTU in patients from northern areas of the continent . It is well known that TcV infection is endemic in Bolivia , Brazil and Argentina but in Colombia is a novel case that requires further investigation; in fact high-resolution markers have been applied to the few isolates of Colombian TcV showing a tailored hybrid profile suggesting a Pan-American import from south America [62] . The DTU TcVI , is mainly detected in the South Cone of Latin America . Normally associated with megavisceral syndromes and some cases of congenital heart disease [4] . In Colombia , TcVI has been very rare and almost infrequent . In fact it is limited to a report in which was detected in humans and R . prolixus isolates ( 4% and 1 . 4% respectively ) . In addition , in different studies with a considerable number of patients conducted in Colombia it was not detected , confirming the low prevalence of the DTU in the country [39 , 51 , 63] . Recently , it has been highlighted the emergence of a genotype named as TcIDom and associated to human infection and domestic transmission cycles via different molecular markers [5 , 6 , 8 , 64–66] . Other studies have shown the presence of TcI Sylvatic genotype in tissue and TcIDom in bloodstream of patients with Chagas cardiomyopathy [41] . In murine models was observed that TcIDom produced high parasitemia and low tissue invasion , a process that allows an adaptation to the host prolonging its permanence and likely generation of chronicity , opposite process to what happened with the TcI sylvatic strains [67] . In accordance with these previous findings , our results show that in chronic patients the frequency and parasitemia of TcIDom genotype were significantly higher in chronic patients than in acute patients , supporting the hypothesis that this genotype may be related to chronicity in patients with Chagas cardiomyopathy . In conclusion , the molecular diagnostic tests are becoming a precise tool to complement the standard diagnostic methods for Chagas disease . This study shows that in general qPCR has a better performance than cPCR . Also , the results confirm that PCR is highly specific for both acute and chronic clinical phases , whereas sensitivity is acceptable for acute phase but still very low for chronic patients . This situation could be partially explained by the higher parasitic loads detected in acute phase and the intermittent nature of the parasite release to the bloodstream in chronic phase . We explored for the first time in a large cohort of Chagas disease patients the DTU parasitemia and the natural course of infection . This type of studies is required in Latin-America for a better understanding of disease progression and molecular epidemiology of Chagas disease . This makes PCR a potential tool for its use in acute phase diagnosis in a routine basis , and potentially for determining aetiological treatment failure when tests positive but not substantially useful when tests negative and these results must be interpreted cautiously as in the clinical trials previously published [21 , 68] . Further research is needed to improve the sensitivity of this test and the mandatory deployment of new diagnostic tests . | Chagas disease is a neglected tropical disease caused by the parasite Trypanosoma cruzi that shows tremendous genetic diversity evinced in at least six Discrete Typing Units and massive genetic diversity within TcI . Two clinical phases exist where acute phase shows high parasitemia and chronic phase shows low and intermittent parasite dynamics . One particularity of the disease is the diagnosis , because the parasitemia is highly variable during the phases of the disease . Molecular tests allow detecting DNA of the parasite in all clinical phases . Herein , we determined the operating characteristics of two molecular tests ( cPCR and qPCR ) to evaluate the performance of these tests for diagnosis of Chagas disease in 708 Colombian patients . We determined the parasitic loads and DTUs to assess how is the behaviour of these characteristics in relation to the clinical phases . We found that the performance of qPCR was higher compared to cPCR and the molecular tests are a precise tool for diagnostic of Chagas disease , mainly in the acute phase . The parasitemia was higher in the acute phase compared to chronic phase and the DTU predominant in Colombian patients was TcI . The behaviour of TcIDom genotype in the chronic phase patients evidenced possible relationship with the chronicity of the disease . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"tropical",
"diseases",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"parasitology",
"parasitemia",
"organisms",
"protozoans",
"neglected",
"tropical",
"diseases",
"molecular",
"biology",
"techniques",
"quantitative",
"parasitology",
"crystallographic",
"techniques",
"research",
"and",
"analysis",
"methods",
"serology",
"artificial",
"gene",
"amplification",
"and",
"extension",
"protozoan",
"infections",
"molecular",
"biology",
"trypanosoma",
"cruzi",
"trypanosoma",
"chagas",
"disease",
"diagnostic",
"medicine",
"biology",
"and",
"life",
"sciences",
"phase",
"determination",
"polymerase",
"chain",
"reaction"
] | 2016 | Molecular Diagnosis of Chagas Disease in Colombia: Parasitic Loads and Discrete Typing Units in Patients from Acute and Chronic Phases |
Retrograde signaling is essential for neuronal growth , function and survival; however , we know little about how signaling endosomes might be directed from synaptic terminals onto retrograde axonal pathways . We have identified Khc-73 , a plus-end directed microtubule motor protein , as a regulator of sorting of endosomes in Drosophila larval motor neurons . The number of synaptic boutons and the amount of neurotransmitter release at the Khc-73 mutant larval neuromuscular junction ( NMJ ) are normal , but we find a significant decrease in the number of presynaptic release sites . This defect in Khc-73 mutant larvae can be genetically enhanced by a partial genetic loss of Bone Morphogenic Protein ( BMP ) signaling or suppressed by activation of BMP signaling in motoneurons . Consistently , activation of BMP signaling that normally enhances the accumulation of phosphorylated form of BMP transcription factor Mad in the nuclei , can be suppressed by genetic removal of Khc-73 . Using a number of assays including live imaging in larval motor neurons , we show that loss of Khc-73 curbs the ability of retrograde-bound endosomes to leave the synaptic area and join the retrograde axonal pathway . Our findings identify Khc-73 as a regulator of endosomal traffic at the synapse and modulator of retrograde BMP signaling in motoneurons .
Bidirectional communication between the neuronal cell body and distant synaptic terminals is essential for synapse formation , plasticity and neuronal survival [1 , 2] . This is achieved primarily through highly regulated axonal transport . Anterograde transport is mediated by plus-end directed kinesin motor proteins that deliver synaptic vesicles and newly synthesized proteins to the synapse , while retrograde transport of cargo destined for the cell body , such as activated receptor complexes , is accomplished by dynein protein complexes [1 , 3–5] . Kinesin and dynein motors are also required for endosomal traffic within the cell . The coordinated action of anterograde and retrograde motors ensures the proper sorting and delivery of signaling complexes , proteins and organelles [6] . Although defects in endosomal traffic and axonal transport have been associated with a number of nervous system diseases including Charcot-Marie-Tooth disease , Amyotrophic Lateral Sclerosis , Huntington’s disease and Parkinson’s disease , we know little about how signaling endosomes are routed from the synapse to the retrograde pathway [7–13] . Retrograde signaling has been extensively studied at the Drosophila larval neuromuscular junction ( NMJ ) . In particular the Bone Morphogenic Protein signaling pathway ( BMP ) has been identified as a major regulator of synaptic growth and function . As such , many regulators of synaptic endosomal sorting have been identified in the regulation of BMP signals at synaptic terminals . Nevertheless , how activated receptors are preferentially sorted to travel to the nucleus is currently unknown . The movement of endosomes within the cytoplasm is directed through the actions of microtubule binding proteins such as minus end dynein motors , and plus end directed kinesins . Co-ordination and competition between these opposing motors for endosome cargoes regulates the transport of proteins to their correct targets [1 , 4 , 14–18] . In this study , we have discovered a surprising role for the plus-end directed microtubule motor protein Khc-73 in retrograde sorting of signaling vesicles at the Drosophila larval NMJ . Khc-73 and its vertebrate homolog KIF13B/GAKIN are kinesin 3 motor protein family members with multiple protein domains and diverse roles in both vertebrates and invertebrates [15 , 19–32] . At its N-terminal , Khc-73 contains a kinesin motor necessary for its association with microtubules and plus-end directed transport to synaptic terminals , and at its C-terminal , a Cytoskeletal Associated Protein GLYcine rich ( CAP-GLY ) domain that provides microtubule association properties [30 , 32] . In the nervous system , through microtubule cytoskeleton interactions , both KIF13B and Khc-73 have been shown to participate in mechanisms that control neuronal polarity: Khc-73 has a role in spindle orientation in neuroblasts [30] , and KIF13B is involved in the establishment of axonal structures in post-mitotic neurons [20] . Interestingly , KIF13B/Khc-73 has been implicated in the regulation of endosomal dynamics [33 , 34] and axonal transport [15] through interaction with Rab5-GTPases . Our previous findings suggested that Khc-73 , under strong inhibitory control of the microRNA miR-310-313 cluster in motoneurons at the Drosophila NMJ , plays an important role in the regulation of synaptic function by influencing presynaptic neurotransmitter release [35] . In order to investigate the mechanism of action of Khc-73 , we have generated loss of function deletions in Khc-73 gene in D . melanogaster and examined the motoneurons of third instar larvae . While the number of synaptic boutons at the NMJ and the amount of neurotransmitter release per action potential are unaffected in Khc-73 mutant larvae , we find a small but significant decrease in the number of presynaptic release sites . Our experiments indicate the presence of Khc-73 function in BMP signaling by demonstrating a strong genetic interaction between Khc-73 and members of the BMP signaling pathway . We further show that activation of retrograde BMP signaling that normally leads to accumulation of pMad in the nuclei of motoneurons is significantly suppressed when Khc-73 is genetically removed . Our findings suggest that Khc-73 exerts its function by influencing the sorting of endosomes at the NMJ and promoting retrograde routing of endosomes . Our findings identify , for the first time , a plus-end directed microtubule motor protein as a regulator of retrograde signaling in motoneurons .
Khc-73 is a member of the KIF superfamily of kinesin motor proteins and the homologue of the vertebrate KIF13B/GAKIN ( S1A Fig ) [30 , 35] . We have previously shown that Khc-73 is a target of the micro RNA miR-310-313 cluster in motoneurons . We found that loss of the miR-310-313 cluster led to abnormally enhanced neurotransmitter release at the NMJ; this enhancement was fully reversed to wild type levels as a result of neuronal knock down of Khc-73 [35] . In order to investigate the role of Khc-73 in more detail , we generated deletions in the Khc-73 gene by imprecise excision of a P-element transposon insert in the vicinity of the 5’ UTR of Khc-73 ( S1B Fig ) . We isolated two deletion flies Khc-73149 and Khc-73193 missing portions of the Khc-73 5’UTR and the ATG start ( S1B Fig ) ; we also isolated a fly where the P-element was excised precisely leaving the entire genetic region of Khc-73 intact ( Khc-73100 ) ( S1B Fig ) . Our western blot analysis with an antibody against the C-terminal end of Khc-73 indicates that both Khc-73149 and Khc-73193 are protein null alleles ( S1C Fig ) . Khc-73 is maternally expressed and is expressed in the embryo [32] . We examined the expression pattern of Khc-73 protein in motoneurons with transgenic overexpression , since we were not able to detect a specific signal using our antibody against Khc-73 in larval preparations . We overexpressed HA-tagged Khc-73 transgene in motor neurons and detected punctate accumulation of HA-Khc-73 both in axons and in synaptic boutons at the NMJ ( S1D and S1E Fig ) . In addition , we tested transcriptional activity of Khc-73 by generating a Khc-73-Gal4 fly ( containing 4kb of Khc-73 genomic sequence driving Gal4 expression , see methods ) . Crossing this fly to UAS-mCD8-GFP transgene led to the expression of GFP in nearly all neurons including motor neurons , suggesting that Khc-73 transcription is active in all motor neurons in third instar larvae ( S1F and S1G Fig ) . We also found Khc-73 transcription widely expressed in the brain of adult flies ( S1H Fig ) . Based on the previously published roles for Khc-73 in neuroblasts , endosomal sorting , axon morphology and synaptic function [21 , 29 , 30 , 34–36] , we expected loss of Khc-73 to cause significant defects in the normal synaptic function and/or structure . To our surprise , we found only mild defects ( S2E and S2F Fig ) in Khc-73 mutant larvae in our assessment of gross synaptic structure at the larval NMJ . The number of synaptic boutons and the muscle surface area ( MSA ) at NMJs were not significantly different comparing Khc-73 mutant and control larvae; this was true for muscle 4 NMJs ( Fig 1A–1C ) as well as muscle 6/7 NMJs ( S1I and S1J Fig ) . To test whether loss of Khc-73 might affect synaptic function , we examined the baseline electrophysiological properties including miniature excitatory postsynaptic currents ( mEPSCs ) , evoked excitatory postsynaptic currents ( EPSCs ) and quantal content ( QC ) and found no differences between Khc-73 mutants and wild type larvae ( Fig 1D and 1E ) . Similarly , we tested synaptic vesicle recycling dynamics in Khc-73 mutant NMJs with high frequency stimulation and found no significant difference in the decay of the synaptic response compared to controls ( S1K and S1L Fig ) . Consistent with the lack of defects in baseline synaptic function , we found no significant changes in the fluorescent intensity of the synaptic vesicle calcium sensor synaptotagmin ( SYT ) , synaptic vesicle marker cysteine string protein ( CSP ) or synaptic vesicle recycling protein Epidermal growth factor receptor pathway substrate clone 15 ( EPS15 ) in Khc-73 mutant larvae ( S2A–S2D Fig ) . We found a mild reduction in the staining intensity for the postsynaptic marker Discs large ( Dlg ) ( S2E and S2F Fig ) but no differences in the expression level of postsynaptic glutamate receptor subunit A ( GluRIIA ) ( S2G Fig ) . Altogether these findings indicate that synaptic growth and function are largely normal in Khc-73 mutant . We have previously reported an increase in the accumulation of the active zone protein Bruchpilot ( Brp ) in miR-310-313 cluster mutant larvae that could be reduced by transgenic knockdown of Khc-73 [35] . Therefore , we set out to conduct a deeper examination of Khc-73 mutants to understand the mechanism of action of Khc-73 in motoneurons . As our previous data would predict , we found a significant decrease in the number of presynaptic release sites per NMJ in Khc-73 mutant larvae , as indicated by a reduction of the number of Brp Puncta ( Fig 2A and 2B ) . Inclusion of a genomic fragment containing the entire genetic region of Khc-73 gene restored synaptic defects in Khc-73 mutant larvae , indicating that this defect is related to loss of Khc-73 ( Fig 2B ) . Previously we showed that Khc-73 is under control of the microRNA cluster miR-310-313 [35] . To maintain this relationship in our tissue specific rescue , we used a Khc-73 transgene Khc-73-3’UTR ( K014 ) [35] that retains this negative regulatory control . We found that transgenic expression of Khc-73 in presynaptic motoneurons , but not in postsynaptic muscles was sufficient to establish a normal number of presynaptic release sites ( Fig 2C and 2D ) . This result indicates that Khc-73 function in motoneurons is required for normal maturation of synaptic release sites . During larval development , both the coordinated growth of synaptic boutons and the establishment of synaptic strength at the NMJ are largely dependent on a retrograde signaling cascade that is initiated by the release of the Bone morphogenic protein Glass bottom boat ( Gbb ) in postsynaptic muscles . Gbb signals through type I and type II BMP receptors , leading to phosphorylation of and subsequent accumulation of the BMP transcription factor Mad ( Mothers against decapentaplegic ) in the nuclei of motor neurons [37–41] . Through this signaling cascade , genes that control synaptic growth and function are transcriptionally regulated [42–44] . The decrease in the number of presynaptic release sites in Khc-73 mutant larvae , therefore , prompted us to examine the state of BMP signaling in these mutants . The first indication of Khc-73 involvement with BMP signaling came from genetic interaction experiments between Khc-73 and the Drosophila homolog of vertebrate SMAD4 , Medea . Medea is a transcriptional co-factor that is required for normal BMP signaling in motor neurons [45] . We found that a combination of previously published alleles MedeaG112 and MedeaC246 resulted in a very small reduction in the number of boutons at the NMJ compared to MedeaC246 homozygous loss of function mutant [45] , suggesting that G112 is a hypomorphic allele ( Fig 3A and 3E ) . Interestingly , in transheterozygous combinations of Khc-73 and Medea , we found a significant reduction in the number of presynaptic release sites per NMJ ( Fig 3A and 3B ) , no change in Brp puncta per bouton ( Fig 3C ) , a significant reduction in synaptic area ( Fig 3D ) and a significant reduction in bouton number with MedC246 allele but not MedG112 ( Fig 3E ) , as compared to heterozygous MedeaG112 controls . This transheterozygous genetic interaction suggested that Khc-73 , while having a mild influence on baseline BMP signaling , becomes critical when BMP signaling is compromised . In support of these results , we also found a strong genetic interaction between Khc-73 mutants and a mutation in the BMP type II receptor wishful thinking ( wit ) : transheterozygous combination between Khc-73 and wit A12 mutants led to a significant reduction in the number of presynaptic release sites ( Fig 3F and 3G ) , synaptic area ( Fig 3I ) and bouton number ( Fig 3J ) but no change in Brp puncta per bouton ( Fig 3H ) . To further explore the functional link between Khc-73 and BMP signaling , we generated double mutant combinations of Khc-73 and Medea . We assessed these double mutant combinations for defects in Brp puncta number at the NMJ and for accumulation of Brp in axons ( as previously reported [46] ) . We found that defects in active zone number and abnormal Brp accumulation in axons in the transallelic combination of MedeaC246/MedeaG112 were not further enhanced upon removal of Khc-73 ( S3A–S3F Fig ) , indicating that Khc-73 and Med likely function in the same and not parallel pathways with respect to these phenotypes . We then tested whether defects in Brp puncta number in Khc-73 mutants can be restored by overexpressing BMP signaling in motoneurons . Indeed , overexpression of a constitutively active form of BMP type I receptor Thick veins ( TKVACT ) in motoneurons was capable of reversing the reduction in Brp puncta defect in Khc-73 mutant larvae ( Fig 4A–4C ) . These results prompted us to compare the degree of axonal accumulation of Brp and another synaptic marker , synaptotagmin ( SYT ) between Khc-73 and Mad mutants . For this we used a Mad mutant allele ( MadK00237 ) that is known to show a strong reduction of synaptic growth and function at the NMJ and exhibit defects in axonal transport of synaptic markers [41] . Both endogenous Brp and SYT accumulated in large aggregates in axons of Mad mutant larvae compared to wild type or Khc-73 mutant larvae , highlighting the fact that Khc-73 related axonal defects would be comparable to a hypomorphic loss of function of BMP signaling ( Fig 5A–5D ) . In support of this interpretation , the increase in Brp accumulation in axons of Khc-73 mutant larvae was fully reversed as a result of overexpression of TKVACT ( Fig 5E and 5F ) . We also tested whether abnormal axonal accumulation of Brp in Khc-73 mutant larvae could be due to changes in microtubule structures; however , we found no significant changes in the expression of acetylated tubulin in axons or terminals in Khc-73 mutant larvae when compared to wild type counterparts ( S3G and S3H Fig ) . These results further support that the defects associated with active zone numbers in Khc-73 mutant larvae are most likely related to defects in BMP signaling . While we did not detect measurable changes in the accumulation of pMad in response to loss of Khc-73 ( S4A and S4B Fig ) , or overexpression of Khc-73 transgene ( S4C and S4D Fig ) , the genetic interactions described above provide strong evidence for a functional link between Khc-73 and BMP signaling . In order to strengthen this link and extend it to the regulation of synaptic function , we conducted a number of electrophysiological examinations . Mild to moderate overexpression of TKVACT in motor neurons can lead to an enhancement in synaptic release without significantly affecting the number of synaptic boutons [44 , 47] . We found that loss of Khc-73 could significantly block the ability of TKVACT to enhance synaptic strength ( Fig 6A and 6B ) . Similarly , we found that overexpression of the BMP ligand Gbb in postsynaptic muscles led to a significant enhancement in quantal content ( Fig 7A and 7B ) . As in the case of TKVACT induced enhancement in neurotransmitter release , loss of Khc-73 led to a significant suppression of Gbb-induced enhancement in neurotransmitter release ( Fig 7A and 7B ) . In addition , we quantified the accumulation of pMad in motoneuron nuclei in the ventral nerve cord ( VNC ) as a result of postsynaptic overexpression of Gbb . It is well accepted that the accumulation of pMad in the nuclei of motor neurons is a reliable readout of the strength and efficiency of retrograde BMP signaling in motor neurons and is essential for BMP-dependent transcriptional regulation as well as regulation of synaptic function [38 , 40 , 44 , 48 , 49] . Muscle overexpression of Gbb led to a statistically significant increase in pMad in the nuclei of motoneurons , which was fully reversed as a result of loss of Khc-73 ( Fig 7C and 7D ) . Finally , we tested whether Khc-73 gain-of-function would be dependent on normal BMP signaling in motoneurons . We have previously shown that Khc-73 overexpression in motoneurons leads to an enhancement of neurotransmitter release [35] . We found that heterozygosity for the BMP type II receptor wishful thinking ( wit ) was sufficient to suppress this enhancement to a large extent ( Fig 8A and 8B ) , further supporting the presence of a functional link between Khc-73 and BMP signaling . From these results a picture emerges , indicating a strong functional link between Khc-73 and BMP signaling in motor neurons . But how does Khc-73 interact with BMP signaling ? BMP signaling in motoneurons depends on tightly regulated endosomal traffic . For example , pMad accumulation in motoneuron nuclei in response to activation of BMP signaling at the synapse is dependent on retrograde routing of signaling endosomes containing BMP receptor complexes from the nerve terminal along axons to the cell body [5] . Conversely , routing of BMP receptor complexes to lysosomal pathways appears as one of the mechanisms that attenuates BMP signaling in motor neurons [50–52] . Therefore , we considered a role for Khc-73 in both retrograde routing as well as lysosomal sorting of BMP receptor complexes . To test these possibilities , we assessed the level of BMP receptors Wit and TKV using a combination of Western blot analysis and immunohistochemistry . Western blot analysis of CNS and body wall muscle tissue ( containing NMJ terminals ) revealed no change in the level of endogenous Wit protein as a result of genetic removal of Khc-73 ( Fig 9A–9D ) . The available antibody to Wit does not detect endogenous Wit in immunohistochemistry . Thus we turned to transgenic Wit and Tkv to visualize their localization at the synapse . Static images of the boutons in live preps of WIT-GFP revealed punctate accumulations at the NMJ and an increase of Wit receptor intensity in Khc-73 mutants at muscle 4 and muscles 6/7 ( Fig 9E–9H ) . Similarly , TKV:YFP transgene expression appeared more punctate at muscle 4 ( S5A Fig ) and muscles 6/7 ( S5C Fig ) , trending towards increased intensity at muscle 4 ( S5B Fig ) , while significantly increasing in intensity at muscles 6/7 ( S5D Fig ) in Khc-73 mutants . We ruled out changes in TKV:YFP transgene transcription by quantitative PCR ( S5E Fig ) and did not observe obvious changes in axonal traffic of TKV:YFP in motoneurons ( S5F Fig ) . We next tested our model that Khc-73 loss can suppress BMP signaling by examining pMAD levels in larvae overexpressing the Wit receptor in presynaptic neurons and in larvae overexpressing the Gbb ligand from postsynaptic muscle . Overexpression of Wit enhanced presynaptic pMad levels ( Fig 10A and 10B ) . In Khc-73 mutants , this enhancement was significantly suppressed ( Fig 10A and 10B ) . Similarly , muscle overexpression of Gbb enhanced pMAD levels in presynaptic boutons . Khc-73 loss also suppressed this increase ( Fig 10C and 10D ) . It has been demonstrated that BMP receptor activity can be dampened when trapped inside the lumen of multivesicular bodies ( MVBs ) at the NMJ [53] . Generally , MVBs are intracellular vesicles that contain one or more smaller vesicles within their lumen and play an important role in signal transduction and endosomal sorting [54 , 55] . Current evidence suggests that MVBs may be at the crossroads for endosomal cargo joining the lysosomal pathway , the retrograde pathway or the exosomal secretory pathway [55 , 56] . We find that fluorescence intensity of the MVB localized protein Hrs ( hepatocyte growth factor related tyrosine kinase substrate ) is increased by 20% at the NMJ in Khc-73 mutant larvae overexpressing the BMP receptor Wit ( S6A and S6B Fig ) . Suggesting that there are more MVBs in Khc-73 mutants in this Wit overexpressing background . Therefore , a scenario can be considered in which retrograde bound BMP receptors are encapsulated in multivesicular bodies and may be stalled at the NMJ in Khc-73 mutants . Together , these results suggest that degradation of BMP receptors is not a likely explanation for the inhibition of BMP signaling in Khc-73 mutant larvae . Secondly , our findings suggest that while BMP receptors appear to accumulate at the NMJs in Khc-73 mutants , they are in an endosomal state that prevents these receptors from signaling . Previous studies on Khc-73/KIF13B have identified endosomal sorting roles for this protein [15 , 21 , 22 , 27 , 28 , 34 , 57] . In order to gain additional insight into the role of Khc-73 in the regulation of endosomal traffic , we conducted an ultrastructural analysis of NMJ synapses in Khc-73 mutant larva . Our analysis revealed no gross abnormalities in presynaptic boutons ( Fig 11A–11F ) : different morphometric measures of active zones and synaptic vesicles appeared normal in Khc-73 mutant larvae ( Fig 11B–11E ) ; however , we did detect a small increase in the depth of the subsynaptic reticulum ( SSR ) ( Fig 11F ) . Interestingly , although we find no statistical difference in the mean MVBs per bouton ( 1 . 28±0 . 29 control and 1 . 68±0 . 41 Khc-73 ) , we found a proportion of boutons with an abnormally higher number of MVBs ( 7–9 MVBs per bouton ) in Khc-73 mutant larvae ( Fig 11G and 11H ) . The trend towards more MVBs in Khc-73 mutant boutons suggested a role for Khc-73 in endosomal sorting . Therefore , we turned to exploring a possible role for Khc-73 in the regulation of endosomal dynamics by examining the expression of transgenic Rab-GTPases at the synapse . Rab-GTPases are small GTPases that associate with endocytic vesicles and are known to mediate many aspects of endosomal traffic in all eukaryotes [58] . Based on previous reports on interaction between Khc-73 with the early endosome associated Rab5 in vitro [34] , we tested the expression pattern of Rab5 at the NMJ in Khc-73 mutant larvae with a UAS-Rab5:YFP transgene . However , we found that in Khc-73 mutants the punctate appearance of Rab5:YFP was unaffected in terms of fluorescence intensity or localization ( Fig 12A and 12B ) . Similarly , we did not detect any effect on the expression level of the recycling endosomal marker Rab11 ( Fig 12C and 12D ) . In most eukaryotic cells Rab5 positive internalized vesicles become associated with Rab7 along their path of maturation [59–62]; Rab7 containing late endosomes are then either routed to the lysosomal pathway or the recycling pathway [58 , 63] . In neurons , the transition from Rab5 to Rab7 is also necessary for routing late endosomes onto the retrograde pathway [64] . The retrograde pathway is necessary for transporting signaling complexes , neurotrophic factors and other cellular proteins from nerve endings to the cell body [2] . Interestingly unlike the case of Rab5 , we found an abnormal increase in Rab7 accumulation at synaptic boutons in Khc-73 mutants ( Fig 12E and 12F ) . These results suggested to us that the normal dynamics of Rab7 positive vesicles and by extension those of BMP receptors are disrupted in Khc-73 mutant larvae . In order to examine the dynamics of late endosomal traffic in more detail , we set out to conduct live imaging in dissected larvae expressing Rab7:GFP . To see if our observations of Rab7:GFP would be relevant to the dynamics of Wit/Tkv complexes , we confirmed in fixed samples that Wit and Rab7:GFP colocalized when expressed simultaneously ( S7A Fig , Pearson’s r coefficient 0 . 68 ) . We also confirmed that Tkv and Wit colocalized at the NMJ ( S7B Fig , Pearson’s r coefficient 0 . 60 ) . In live dissected larval preparations , Rab7:GFP showed dynamic movement within synaptic boutons in both wild type and Khc-73 mutants ( Fig 13A–13C and S1 Movie and S2 Movie ) . We noticed that occasionally a Rab7 marked vesicle left the synaptic area and moved retrograde towards the shaft of the axon . Vesicles entering the axon moved , paused and continued moving out of the NMJ . We measured the velocity of these vesicles when in motion and calculated the mean velocity in the anterograde and retrograde directions ( Fig 13B–13E , S8A–S8D Fig and S3 Movie and S4 Movie ) and found no statistical difference in their velocities . We also recorded the time spent paused in a single spot ( Fig 13F ) , the number of pauses for each spot ( Fig 13G ) and summed the total time paused in the axon ( Fig 13H ) . Here , our assessment of Rab7 dynamics revealed a significant difference between control and Khc-73 mutant larvae . We recorded long periods of pausing or stalling of Rab7 positive vesicles in Khc-73 mutants , which showed statistical difference compared to our recordings in control larvae ( Fig 13F and 13H , S1 Movie and S2 Movie ) . This pausing phenotype provides one explanation for the increase in Rab7:GFP in Khc-73 NMJs , however alternative explanations related to Rab7:GFP protein turnover are also possible . We next performed time lapse imaging on TKV-YFP expressing Khc-73 mutant larvae focusing on the axon shaft near the synapse . Here we observed a similar stalling phenotype of TKV-YFP puncta in Khc-73 mutants whereas in control larvae the axonal shaft was devoid of stalled puncta ( Fig 13I and 13J , S5 Movie and S6 Movie ) ) . As an additional test for axonal retrograde transport , we used a peripheral axon injury model developed by Collins and colleagues for activating Jun-N-terminal kinase ( JNK ) signaling in motor neurons [65] . In this model , crushing peripheral axons in larvae leads to a strong transcriptional upregulation of the JNK phosphatase puckered ( puc ) in the injured motoneurons [65] . The puc transcriptional response to axon injury is dependent on axonal retrograde transport [65] . Using a puc-LacZ transcriptional reporter line , we assessed JNK activation in motoneurons in response to nerve crush . In Khc-73 larvae , we found that puc transcriptional upregulation as a result of axonal injury was indistinguishable from that of control larvae ( S8E–S8H Fig ) . Thus we can rule out a defect in retrograde axonal transport in Khc-73 mutants . Similarly , we did not find any significant changes in axonal transport of mitochondria in Khc-73 mutant larvae ( S7 Movie and S8 Movie ) . These results provided strong evidence for a model in which Khc-73 is required primarily in synaptic terminals for efficient routing of retrograde vesicles onto the retrograde path with little influence on bidirectional axonal transport .
Khc-73 function plays a supporting role in retrograde BMP signaling under basal conditions . However under conditions of enhanced BMP signaling , this endosomal coordination by Khc-73 becomes critical to transmit the retrograde signal from the synapse to the neuronal cell body . Efficient retrograde signaling from synaptic terminals back to the neuronal soma is critical for appropriate neuronal function and survival [2 , 7–11] . Nevertheless , we know very little about the molecular steps that facilitate the routing of synaptic endosomes destined for retrograde axonal pathways . Here we describe several lines of evidence for a potential role for Khc-73 in this process . Khc-73 mutant larvae develop grossly normal synaptic structure and function at the Drosophila larval neuromuscular junction ( NMJ ) , but we find a reduction in the number of presynaptic release sites . Through genetic interaction experiments , we show that this defect is most likely the result of abnormal BMP signaling in motoneurons: transheterozygous combinations of Khc-73 and Medea or wit mutants show a significant loss of presynaptic release sites compared to control . Khc-73 becomes even more critical , when higher demand is put on the motoneuron by activating BMP signaling: loss of Khc-73 largely blocks the retrograde enhancement in synaptic release in response to activation of BMP pathway in motor neurons . Consistently we have previously shown that transgenic knock down of Khc-73 in motoneurons blocks the ability of the NMJ to undergo retrograde synaptic homeostatic compensation [35] . Our findings show that when BMP signaling is activated , loss of Khc-73 reduces the accumulation of pMad in motoneuron nuclei , suggesting a role for Khc-73 in the regulation of retrograde signaling . Our immunohistochemical assessment and live imaging analysis of Khc-73 mutant larvae provide evidence for involvement of Khc-73 in at least two steps in endosomal dynamics in motoneurons . On the one hand , Khc-73 is required for normal dynamics of internalized endosomes through late endosomal and multivesicular stages , and on the other Khc-73 plays a role in facilitating the routing of endosomes onto the retrograde pathway ( see Fig 14A for model ) . These defects have two main consequences: first , we find an accumulation of BMP receptors at the NMJ ( possibly in multivesicular bodies ) without increased local signaling , suggesting that these receptor containing endosomes might be trapped in a state between late endosomal and lysosomal stage ( see Fig 14B for model ) . Second , we see a dampening of the ability of retrograde bound Rab7:GFP tagged endosomes to join the retrograde pathway , illustrating a defect in retrograde movement of vesicles and possibly providing an underlying explanation for the reduction in pMAD when retrograde BMP signaling is activated in Khc-73 mutants . These results together present Khc-73 , a plus-end microtubule motor , in the unexpected role of regulation of endosomal traffic from synapse to the soma in motoneurons with a role for ensuring the efficiency of retrograde BMP signaling . While our findings provide compelling evidence for the proposed model above , we cannot , at this time , rule out the possibility that the abnormal accumulation of BMP receptors at the NMJ and the slowing of retrograde movement of Rab7 positive endosome in Khc-73 mutant larvae could be due to a defect in an intermediate molecule , whose anterograde transport is dependent on Khc-73 . In support of such model , we do report an abnormal accumulation of Brp and SYT ( two synaptic proteins ) in axons . While our data suggests that this abnormal accumulation can be remedied by transgenic activation of BMP signaling in Khc-73 mutants , we cannot rule out the possibility that an anterograde transport defect might exist for other proteins independent of the interaction between Khc-73 and BMP signaling . Our findings point to a model in which Khc-73 facilitates the routing of retrograde bound vesicles onto the retrograde axonal pathway . This model predicts coordination between endosomes , dynein motors and kinesin Khc-73 . The coordinated involvement of dynein and kinesin motor proteins in the transport and sorting of endosomes has been previously proposed and examples supporting this model are mounting [14 , 66 , 67] . Previously published data for Khc-73 and KIF13B have provided evidence that interaction between early endosomes , dynein motors and microtubules are possible . Khc-73/KIF13B is capable of binding to the GTPase Rab5 ( found on early endosomes ) , thus allowing Khc-73 to localize directly to Rab5 endosomes [15 , 34 , 68] . As a kinesin motor protein , Khc-73 could then transport these endosomes to the retrograde pathway by moving along the microtubule network in the synapse . Compelling evidence for a dynein interaction with Khc-73 has been previously demonstrated during mitotic spindle formation [24] . The Khc-73/KIF13B stalk domain is phosphorylated by Par1b and this creates a 14-3-3 adapter protein binding motif [29] . It has been proposed that physical interaction between Khc-73 stalk domain and the dynein interacting protein NudE via 14-3-3 ε/ζ might underlie the interaction between Khc-73 and dynein that is necessary for appropriate mitotic spindle formation [24] . Interestingly , transgenic knock down of NudE in Drosophila larval motoneurons leads to a reduction in the number of presynaptic release sites , a phenotype reminiscent of Khc-73 loss of function [69] . Thus , Khc-73 contains domains and protein-protein interactions that are capable of coordinating endosomes , microtubules and dynein . We propose that Khc-73 is necessary for the normal endosomal sorting and exit of endosomes from the NMJ to support efficient retrograde BMP signaling .
Flies were cultured at 25°C on standard medium except for Gene Switch experiments where RU486 was added to the media ( 50μM ) . The following stocks were used: MedC246 ( Y324term mutation ) [45] and MedG112 ( mutation in splice donor site of exon 4 ) [45] from Herman Aberle [45] . witA12 [37 , 39 , 70] . witHA4 [37 , 45 , 49] . UAS-TKVACT and UAS-Gbb99 [40] provided by M . B . O’Connor ( University of Minnesota , Minneapolis , MN ) , UAS-Wit [39] , UAS-HA-Khc-73 and UAS-HA-Khc-73-3’UTR ( K014 ) [35] , UAS-Wit-GFP , UAS-TKV-YFP [5] , BG380-Gal4 [71] , Elav-Gal4 [72] , OK371-Gal4[73] , MHC-Gal4 [74] . Bloomington stocks used were P{y[+m8] = Mae-UAS . 6 . 11}Khc-73[DP00530] ( RRID:BDSC_22058 ) , UAS-Rab5:YFP ( RRID:BDSC_9775 ) , UAS-Rab7:GFP ( RRID:BDSC_42706 ) , UAS-Rab11:GFP ( RRID:BDSC_50782 ) , VGlut-Gal4 ( RRID:BDSC_24635 ) , MadK00237 ( RRID:BDSC_10474 ) , UAS-Mito-HA-GFP ( RRID:BDSC_8442 ) , nSyb-Gal4 ( RRID:BDSC_51635 ) . UAS-luciferase ( RRID:BDSC_35788 ) . puckered LacZ insertion pucE69 [75] . Wild type stock used was w1118 . Khc-73 deletions were created by mobilizing the P-element from y[1] w[67c23]; P{y[+m8] = Mae-UAS . 6 . 11}Khc-73[DP00530] . Virgin y[1] w[67c23]; P{y[+m8] = Mae-UAS . 6 . 11}Khc-73[DP00530] female flies were mated to Cyo/+; Δ2–3 , Sb/TM6b males . Male progenies of y[1] w[67c23]; P{y[+m8] = Mae-UAS . 6 . 11}Khc-73[DP00530]/Cyo; Δ2–3 , Sb/+ were mated to virgin y , w; CyoGFP/Adv females . Yellow , non Sb , yellow eyed progeny were singly mated to y , w; Adv/CyoGFP virgins and individual stocks were established . P-element excisions were screened with the following primers: OED91: CTGACGGCGCTGTTGCTTG and OED96: GATCTAGAGATGATTCTGCATCACTAG TAAAAATT . Khc-73 promoter GAL4 construct was generated by cloning a 4kb fragment upstream of the translational start site with primers OED453: CAG GTA CCG CCG AGG AAC CGC TAA CG and OED452:CAG GTA CCC GCG GAT GTG GAT GCA GC . Vector pW+SN attB was modified with a GAL4 sequence cloned as a KpnI/NotI fragment . Khc-73 promoter was subsequently inserted into the unique KpnI site . Genomic Khc-73 is from BACPAC clone CH321-36I16 ( BACPAC Resources Center ) . Transgenic fly CH321-36I16 was made by standard embryo injection of BACPAC clone CH321-36I16 ( BACPAC Resources Center ) with ΦC31 –mediated integration into attP site at position 86F of chromosome III . Wandering third instar larvae were dissected , prepared and embedded as described in [76] . Ultra-thin serial sections of 50 nm thickness were cut from muscle 6 , 7 and 12 of hemisegment A3 . Electron micrographs were taken at a magnification of 25 , 000x for measurements , 25 , 000x and 40 , 000x for figures . Serial Reconstruction and analysis was conducted on FIJI ( Fiji is Just ImageJ ) ( NIH ) [77] and Reconstruct v . 1 . 1 . 0 . 0 Software [78] . Wandering third instar larvae were dissected as previously described [74] . Third Instar larvae were dissected in cold HL3 and fixed with 4% Paraformaldehyde for 10 min or 5min ice cold Methanol for GluRIIA staining . Larvae were washed with PBS ( Phosphate buffered saline ) , permeabilized with PBT ( PBS with 0 . 1% Triton X-100 ) , blocked with 5% Normal Goat Serum ( NGS ) in PBT and placed in primary antibody overnight at 4°C . The larvae were then washed three times for 15min in PBT , placed in secondary antibody for 2 hrs , washed three times for 15min with PBT and mounted in Vectashield ( Vector labs ) . Antibodies used are as follows: anti-GluRIII ( 1:500 ) ( gift from A . DiAntonio , Washington Univ . St . Louis , MO ) , anti-Hrs ( 1:200 ) , anti-SYT ( 1:1000 ) ( gift from H . Bellen , Baylor College of Medicine , Houston , TX ) , anti-pMAD ( PS1 ) ( 1:200 ) ( gift from M . B . O’Connor , University of Minnesota , Minneapolis , MN ) . anti-Dlg ( 1:500 ) , anti-nc82 ( 1:500 ) , anti-GluRIIA ( 1:500 ) , anti-CSP ( 1:500 ) , anti-EPS15 ( 1:50 ) , anti-LacZ ( 1:100 ) and anti-Wit ( 1:10 ) ( Developmental Studies Hybridoma Bank ( DSHB ) ) , anti-HA ( 1:500 ) ( HA . 11 clone 16B12 ) ( Covance Research Products ) , anti-GFP ( 1:500 ) ( A6455 ) ( Molecular Probes ) , anti-GFP ( 1:500 ) ( Rat IgG2a , GF090R ) ( Nacalai Tesque Inc . ) , anti-HRP conjugated Alexa 647 ( 1:250 ) ( Jackson ImmunoResearch ) , anti-acetylated tubulin ( 1:500 ) ( T7451 , clone 6-11B-1 Sigma-Aldrich ) and anti-pSmad3 ( EP823Y ) ( Epitomics ) . Western blots were performed as previously described [41] . Muscle tissue ( without the nervous system and motor axons or imaginal discs ) or Brain tissue ( VNC and axons ) were isolated from wandering third instar larvae dissected in cold HL3 . Western blot analysis was performed according to manufacturer’s protocols . Antibodies used: anti-Khc-73 ( 1:2000 ) [35] , anti-Wit ( 1:10 ) ( DSHB ) , anti-actin ( Millipore , MAB1501 ) . Gel images were scanned and band intensities were quantified using FIJI ( Fiji is just ImageJ software ) ( NIH ) [77] . Synapses were imaged using a ConfoCor LSM710 and Zeiss LSM 780 on an Axiovert 200M inverted microscope ( Carl Zeiss , Inc . ) with 63x/1 . 4 oil objective . Image analysis was performed with ImageJ 1 . 46j ( NIH ) [79] , Imaris ( Bitplane Scientific Software ) , Image Analyst MKII ( Image Analyst Software , Novato , CA ) and Metamorph ( Molecular Devices ) . Wandering third instar larvae were dissected in room temperature HL3 to remove the guts and fat bodies . The larval filet was then inverted and stretched onto a coverslip using magnetic dissection pins inside a chamber consisting of a coverslip surrounded by magnet strips . Larval prep was maintained at room temperature in an HL3 bath during imaging . NMJs at hemisegment A3 and A4 , muscles 6/7 and 4 were imaged . Axons were imaged at hemisegment A3 to A4 . Larvae were imaged for a maximum of 30 minutes after dissection . Axons and NMJs were imaged with 63x 1 . 4NA oil objective on Axiovert 200 inverted microscope with Zeiss LSM780 confocal ( Carl Zeiss , Inc . ) . The nerve crush assay was performed as previously described [65] . Briefly , third instar larvae were anaesthetized with carbon dioxide . The segmental nerves at the midbody were then pinched with Dumostar number 5 forceps for five seconds . The larvae were then recovered on standard media for 25 hours at 25°C after which time they were dissected and stained for LacZ . Wandering third instar larvae were dissected in cold HL3 solution following standard protocol [80] . The spontaneous ( mEJC ) and evoked ( EJC ) membrane currents were recorded from muscle 6 in abdominal segment A3 with standard two-electrode voltage-clamp technique [41] . All the recordings were performed at room temperature in HL3 solution containing 0 . 5mM Ca2+ unless otherwise indicated . The current recordings were collected with AxoClamp2B amplifier ( Molecular Devices Inc . ) using Clampex 9 . 2 software ( Molecular Devices Inc . ) . The nerve stimulation was delivered through a suction electrode which held the cut nerve terminal cord . In all voltage clamp recordings , muscles were held at -80 mV . The holding current was less than 5 nA for 90% of the recordings and we rejected any recording that required more than 10 nA current to maintain the holding potential . The amplitudes of mEJC and EJC were measured using Mini Analysis 6 . 0 . 3 software ( Synaptosoft ) and verified by eye . QC was calculated by dividing the mean EJC amplitude by mean mEJC amplitude . The recording traces were generated with Origin 7 . 5 software ( Origin Lab ) . Spontaneous and evoked potentials were measured as previously described [49] . Standard two-electrode voltage-clamp technique was used as described in [44] . | Retrograde axonal transport is essential for normal synaptic function and neuronal survival . How endosomes are specifically sorted from the synaptic terminal for retrograde axonal transport is currently not known . At the Drosophila neuromuscular junction , receptors for the Bone Morphogenic Protein signaling pathway are transported from the synapse to the neuron cell body for the proper establishment of synaptic growth and function of motoneurons . Using this system we demonstrate that a kinesin motor protein , Khc-73 , is necessary for the efficient sorting of retrograde bound vesicles to the retrograde transport route . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"medicine",
"and",
"health",
"sciences",
"vesicles",
"nervous",
"system",
"cell",
"processes",
"electrophysiology",
"neuroscience",
"motor",
"neurons",
"developmental",
"biology",
"molecular",
"motors",
"nerve",
"fibers",
"cellular",
"structures",
"and",
"organelles",
"motor",
"proteins",
"endosomes",
"animal",
"cells",
"axons",
"proteins",
"life",
"cycles",
"axonal",
"transport",
"biochemistry",
"signal",
"transduction",
"cellular",
"neuroscience",
"cell",
"biology",
"anatomy",
"synapses",
"neurophysiology",
"physiology",
"neurons",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"cell",
"signaling",
"larvae",
"bmp",
"signaling"
] | 2018 | Kinesin Khc-73/KIF13B modulates retrograde BMP signaling by influencing endosomal dynamics at the Drosophila neuromuscular junction |
Reduced insulin/IGF signaling increases lifespan in many animals . To understand how insulin/IGF mediates lifespan in Drosophila , we performed chromatin immunoprecipitation-sequencing analysis with the insulin/IGF regulated transcription factor dFOXO in long-lived insulin/IGF signaling genotypes . Dawdle , an Activin ligand , is bound and repressed by dFOXO when reduced insulin/IGF extends lifespan . Reduced Activin signaling improves performance and protein homeostasis in muscles of aged flies . Activin signaling through the Smad binding element inhibits the transcription of Autophagy-specific gene 8a ( Atg8a ) within muscle , a factor controlling the rate of autophagy . Expression of Atg8a within muscle is sufficient to increase lifespan . These data reveal how insulin signaling can regulate aging through control of Activin signaling that in turn controls autophagy , representing a potentially conserved molecular basis for longevity assurance . While reduced Activin within muscle autonomously retards functional aging of this tissue , these effects in muscle also reduce secretion of insulin-like peptides at a distance from the brain . Reduced insulin secretion from the brain may subsequently reinforce longevity assurance through decreased systemic insulin/IGF signaling .
Reduced insulin/IGF-1 signaling increases the lifespan of nematodes , flies and rodents [1] , [2] . In Caenorhabditis elegans , mutants in insulin-like receptor daf-2 live twice as long as wild type [3] , [4] . Mutation of insulin receptor InR and insulin receptor substrate ( chico ) increase adult lifespan in the fruit fly Drosophila melanogaster [5] , [6] . It is reported that mice with mutation at the IGF-1 receptor ( Igf1r ) extend lifespan [7] , as do mutants of the insulin receptor substrate ( Irs2 ) [8] and of the insulin receptor within adipose tissues [9] . Genetic evidence places the forkhead transcription factor FOXO as the downstream effector of insulin/IGF-1 signaling [3] , [10] , [11] , [12] . Activated insulin/IGF-1 signaling enhances the phosphorylation of FOXO , which is sequestered in the cytoplasm . Conversely , reduced insulin results in FOXO nuclear translocation , which thus promotes or represses the transcription of FOXO target genes [11] ( Figure S1A ) . In C . elegans lifespan extension of daf-2 and age-1 ( PI3 kinase ) mutants requires daf-16 , a FOXO homolog in worms [3] . Recent work likewise shows that FOXO is required for insulin-mediated lifespan extension in Drosophila [13] , [14] . FOXO also appears to function in human aging where independent studies found polymorphisms of FoxO3A to associate with exceptional longevity [15] , [16] . Insulin signaling through its control of FOXO is a potentially conserved system to regulate aging but despite this emerging consensus , the proximal targets of insulin/FOXO signaling that orchestrate these mechanisms of longevity assurance are essentially unknown . One aspect that is clear is that insulin/FOXO signaling operates both nonautonomously and autonomously to control Drosophila aging . Systemically reducing insulin signaling by mutations of the insulin receptor ( InR ) and insulin receptor substrate ( chico ) slows the decline in cardiac performance of aging flies while a similar outcome is produced by overexpressing FOXO and PTEN just within cardiomyocytes [17] . Likewise , overexpressing FOXO in muscle maintains muscle protein homeostasis and delays muscle function decline with age while dFOXO expressed in muscles extends lifespan [18] , as does expression of dFOXO only from fat body [19] , [20] . dFOXO expressed from fat body reduces secretion of systemic insulin-like peptides ( DILP2 and DILP5 ) , which are produced predominantly in the brain . dFOXO of Drosophila fat body modulates lifespan by inducing fat body dilp6 transcription , which in turn suppresses neuronal DILP secretion [21] . These findings suggest that insulin/FOXO signaling within some organs controls both the systemic level of circulating DILPs while systemic DILPs regulate somatic maintenance of insulin sensitive tissues . Identifying the FOXO target genes and somatic maintenance pathways in such tissues will elucidate how reduced insulin/IGF-1 assures longevity . Genome-wide studies with C . elegans have been used to probe how daf-16 controls lifespan in response to insulin signaling . Microarray analyses have identified many mRNA that are affected directly or indirectly by daf-16 , and reducing some of these genes by RNA interference ( RNAi ) increases longevity [22] , [23] . Chromatin immunoprecipitation ( ChIP ) and DNA adenine methyltransferase identification ( DamID ) have been used to identify the direct targets of DAF-16 to clarify which pathways are proximally responsible for the impact of daf-16 upon aging [24] , [25] . Oh et al . [24] thus described 103 genes to be direct targets of daf-16 in the long-lived daf-2 ( e1370 ) mutant , and three out of 33 tested target genes were found to increase lifespan when gene expressions were reduced by RNAi: lin-2 , egl-10 and sca-1 . In Drosophila , work to date has identified binding targets of dFOXO primarily from wild type adults . Alic et al identified 1423 dFOXO binding sites in wildtype adult female Drosophila using a ChIP-on-chip approach [26] . Among 1755 unique genes that are less than 1 kb away from these dFOXO binding sites , about 365 genes are transcriptionally regulated by dFOXO . These targets could potentially modulate many insulin related phenotypes including growth , reproduction , and metabolism , as well as aging . Here we aim to understand how reduced insulin/IGF-1 signaling extends Drosophila lifespan by identifying genes transcriptionally regulated by dFOXO in long-lived insulin signaling mutants . We conducted ChIP analysis with a dFOXO antibody followed by Illumina high-throughput sequencing from chico heterozygous mutants , which are long-lived and normal sized , and from adult flies with ablated insulin producing cells ( IPCs ) , which are also long-lived [27] . dFOXO was seen to bind at promoters of 273 genes common to these genotypes , thus providing a candidate set of potential factors in control of aging . Pathways enriched within this set include those for G-proteins , Wnt and Transforming growth factor-beta ( TGF-β ) . We subsequently focused on TGF-β signaling because dFOXO binds in the promoter and represses dawdle ( daw ) , an Activin-like ligand in TGF-β superfamily . In genetic trials , reducing daw or its downstream transcription factor Smox increases lifespan , preserves muscle function and reduces poly-ubiquitinated protein accumulation . The muscle specific benefits of activated dFOXO are mediated through the control of autophagy by Smox , which we find to bind and transcriptionally repress Atg8a/LC3 , a reported longevity assurance gene of Drosophila [28] . Expressing Atg8a in muscle was also sufficient to increase lifespan . Furthermore , reducing daw in muscle decreased DILP2 peptide secretion from the brain while peripheral insulin/IGF signaling was correspondingly reduced . Our results suggest that insulin/IGF signaling controls lifespan in part through dFOXO-mediated repression of muscle Activin signaling and its downstream functions including muscle autophagy , muscle proteostasis and subsequent remote control of systemic insulin/IGF signaling .
To understand how dFOXO extends Drosophila lifespan we sequenced promoters derived from chromatin-immunoprecipitation with antibody against dFOXO in two genotypes of long-lived flies with reduced insulin signaling . Heterozygotes of chico1 live 36% longer than co-segregating wildtype sibs [13] , [29] . Unlike many mutants of the insulin-signaling pathway , chico+/− adults have normal development time , body size and fecundity . Aging is likewise retarded by partially ablating adult IPCs by inducing apoptosis with a cell specific inducible driver ( Dilp2-GeneSwitch-gal4>UAS-reaper ) [27] . We conducted ChIP-Seq analysis from 15-day old female adults from both genetic manipulations . This revealed dFOXO to bind at 1331 and 763 promoter regions ( Figure S1B ) , from chico and IPCs ablated flies respectively , corresponding to 2042 and 1012 candidate genes ( Figure 1A and Table S7 ) . We identified 273 genes common to both longevity-assurance genotypes ( Figure 1A ) . Biological functions defined by Gene Ontology ( david . abcc . ncifcrf . gov ) in this overlapping set include development , growth and neuron differentiation ( Figure 1B ) . Pathway analysis ( david . abcc . ncifcrf . gov ) revealed enrichment in Wnt and TGF-β signaling ( Figure S1C ) . Corresponding to previous work , we also found significant binding of dFOXO at puckered ( puc ) in both longevity assurance genotypes ( Table S1 ) . In the JNK signaling pathway , puc is a negative regulator of JUN kinase basket ( bsk ) , and mutation of puc extends Drosophila lifespan [30] . To determine how candidate dFOXO targets affect longevity we selected 23 genes for further analyses ( Figures 1C–1E , Table S1 ) based on their placement in recognized signaling pathways or because they showed a strong dFOXO binding . dFOXO binding at the promoter of these candidates was verified by ChIP followed with gene specific qPCR . In this analysis dFOXO was significantly enriched at all candidate targets in both insulin mutants when compared to wildtype ( Figures S1E–S1F ) . To measure the impact of insulin/IGF-1 on candidate transcription we quantified mRNA in adults of wildtype ( WT ) , chico null mutant ( chico −/− ) and chico; foxo double mutant ( chico −/−; foxo −/− ) ( Figures 1C–1E ) . Transcripts of 12 genes were up-regulated in chico −/− relative to wildtype but not in chico −/−; foxo −/− , indicating that dFOXO induces these genes . The expression of seven genes was repressed in chico −/− relative to wildtype but not in chico −/−; foxo −/− , suggesting that dFOXO directly represses these genes . Four genes were not differentially expressed despite their enriched dFOXO binding in the insulin mutants; activated dFOXO may be required but not sufficient to control the expression of these genes . Thus , dFOXO can function as both a transcriptional activator and repressor [26] , but this factor may also become poised at genes upon reduced insulin signaling and not affect transcriptional changes until the required co-factors are induced by other signals . To determine if candidate dFOXO targets contribute to aging regulation we measured lifespan and age-specific mortality when each was reduced by RNAi or over-expressed from transgenes . Cohorts of control and mis-expression genotypes were coisogenic; misexpression was induced only in adults via GeneSwitch ( GS ) -Gal4 driving either UAS-RNAi or UAS-transgene . The effect of RNAi on lifespan was assessed for all 23 candidates . Knockdown of three genes ( daw , Glyp and Tsp42Ef ) extended lifespan by consistently reducing age-specific mortality ( Figures 1F–1H , Figure S2 and Table S1 ) , while knockdown of 14 genes shortened lifespan ( Table S1 ) . Among the candidates whose transcriptions were positively regulated by dFOXO , seven transgenic lines were available to test the effect of overexpression on lifespan; two cases had no effect on lifespan while five cases reduced survival ( Table S2 ) . Among the observed longevity assurance genes , daw-RNAi induced by two independent ubiquitous GeneSwitch drivers respectively extended lifespan 12% to 35% ( mean lifespan ) by consistently reducing mortality rate ( Table S1 , S3 ) . Dawdle is one of two Drosophila Activin-like ligands [31] , [32] , belonging to the Transforming Growth Factor-β ( TGF-β ) protein superfamily . To date , daw is reported to function in axon guidance [31] , [32] , cell proliferation and larval brain development [33] . Our results indicate that daw acts as a downstream target of dFOXO to modulate lifespan , suggesting that the Activin branch of TGF- β signaling may participate in control of aging . Drosophila has two TGF-β ligand subfamilies: bone morphogenetic proteins ( BMP ) ( ligands: Dpp , Gbb and Scw ) and Activin ( ligands: Daw and Act-β ) . These ligands signal through subfamily-specific Type I receptors ( Tkv and Sax for BMP , Babo for Activin ) and shared Type II receptors ( Punt , Wit ) . BMP-like ligands and Activin-like ligands activate distinct downstream signaling cascades leading respectively to phosphorylation of the Smad transcription factors Mad and Smox ( Figure 2A ) [34] . Since daw-RNAi increases lifespan , we determined whether other elements of either TGF-β pathway could likewise control aging ( Figures 2C–2F and Figure S3 ) . RNAi for Smox , the Activin associated Smad transcription factor , extended lifespan 10% ( Figure 2F ) . RNAi for Activin receptor babo and the Activin-like ligand Act-β did not affect survival . Repressing the BMP branch of TGF-β signaling via RNAi for dpp , gbb , Mad and Tkv consistently reduced survival ( Table S3 ) . Ubiquitously overexpressing genes in either Activin or BMP subfamily shortened lifespan ( data not shown ) , as did RNAi for co-Smad ( Med ) , the shared Type-II receptor ( Punt and Wit ) and two other TGF- β ligands ( Myo and Mav ) ( Table S3 ) . The TGF-β signaling pathways of Drosophila are homologous to C . elegans TGF-β/dauer and Sma/Mab . Recent reports clarify that the TGF-β/dauer pathway can regulate somatic aging , while the Sma/Mab pathway appears to modulate reproductive aging [35] , [36] . We performed a phylogenetic analysis on TGF-β ligands of C . elegans , Drosophila and mouse ( Figure 2B ) . Similar to previous published phylogenetic analysis [37] , [38] , we found that the TGF-β/dauer ligand of C . elegans , DAF-7 , is closely related to Activin-like ligands in Drosophila ( Daw and Activin-β ) and mouse ( Activin-A , B , C and E ) , while the Sma/Mab ligand in C . elegans , DBL-1 , is similar to BMP-like ligands in Drosophila and mouse . Together these results suggest that Activin may be a conserved longevity pathway . To understand how Activin regulates Drosophila aging we determined which tissues produced this control . daw mRNA is highly expressed in muscle and fat body , a tissue with both liver and adipose-like activities ( Figure 3A ) . Smox protein is more widely distributed ( Figure 3B ) . To assess the role of Activin in muscle and fat body we knocked down daw , Smox and babo with tissue-specific drivers . Lifespan was extended by inactivating each of these genes in muscle , but not in fat body ( Figures 3C–3H , Figure S4 and Table S4 ) . Since daw is a dFOXO downstream target that is down-regulated in chico mutants ( Figure 1D ) , we determined whether reduced insulin/IGF-1 signaling modulates Activin within muscle . chico mutants showed reduced daw mRNA sampled from thorax ( containing mostly flight muscle ) , and this effect was reversed in chico; foxo double mutants ( Figure 3I ) . Furthermore , Smox protein was less phosophorylated in chico mutants ( Figure 3J ) . Insulin/IGF-1 signaling , through dFOXO , thus appears to modulate muscle Activin signaling , which in turn is sufficient to regulate longevity . Muscle performance in many animals declines in parallel to the accumulation of misfolded protein aggregates [39] . Insulin/IGF-1 signaling in Drosophila may affect this process since over-expressing dFOXO in Drosophila muscle slows the aggregate accumulation and promotes macroautophagy [18] . Here we determine whether dFOXO mediates its effects on muscle proteostasis and function through its control of Activin . Experimentally reducing Activin prevents the decline of muscle function with age . Flight activity typically declines in aging flies [18] , as it does in our wildtype control ( Figures 4A–4C ) . RNAi against the Activin factors daw , Smox and babo each retarded this decline ( Figures 4A–4C ) . Likewise , the ability to climb at advanced ages was preserved in daw RNAi flies relative to wildtype ( Figure S6D ) . Progression of these composite movement traits was associated with changes in protein aggregates within muscle . Aggregates visualized with Poly-Ubiquitin FK2 antibody increase with age in wildtype muscle , but this change was significantly delayed by muscle specific RNAi against daw , Smox or babo ( Figure 4D , F ) . Macroautophagy modulates protein aggregate accumulation [40] . We used two markers of lysosome/autophagy activity , lysotracker and cherry-tagged-Atg8a ( homolog of LC3 ) , to determine if Activin regulates muscle proteostasis through macroautophagy . The intensity of lysosome markers decreased with age in wildtype flight muscle , but was maintained in aged muscle expressing RNAi for daw , Smox , or babo ( Figure 4E , quantified in Figure 4G ) . We likewise observed more autophagosomes in flight muscle with inactivated TGF- β/Activin signaling ( via RNAi for daw , Smox , or babo ) ( Figure 5A ) . In contrast , constitutively activated Activin signaling ( via overexpressing babo-Act ) reduced the number of autophagosomes ( Figure 5A , 5B ) . Since Activin signaling is transcriptionally regulated by dFOXO via daw , these results may explain reported associations between reduced insulin signaling and elevated autophagy [18] . Reduced insulin signaling represses Activin , which in turn releases repression of autophagy and thereby reduces accumulation of protein aggregates . Drosophila encode 18 autophagy genes [41] . Many of these are less expressed in aged flies ( Figure 5C ) [18] . Since reduced Smox mRNA produces elevated autophagy in aged muscle , we studied the phosphorylation of this transcription factor in old flies . Smox phosphorylation is increased in aging muscle ( Figure 5D ) , suggesting that Activin may be a negative regulator of Atg gene expression . Indeed , Atg6 and Atg8a mRNA were increased when daw and Smox were reduced in muscle ( Figure 5E , 5F ) , while mRNA of Atg5 , Atg6 and Atg8a were reduced by over-expressing constitutively active form of the babo receptor ( Figure 5G ) . Drosophila Smox protein is homologous to vertebrate Smad2 and Smad3 transcription factors . Human Smad3 protein recognizes the consensus sequence GTCTAGAC [42] , although a single copy of the Smad box ( GTCT ) is also reported to support Smad3 binding at the MH1 domain [43] . We searched the promoter regions of Atg8a and identified at least two adjacent Smad boxes located within Atg8a ( Figure 6A ) . ChIP-PCR with affinity-purified Smox antibody showed that Smox binds to the promoter region of Atg8a ( Figure 6B ) , but not Atg1 and Atg6 ( Figure 6C ) . In contrast to Smox , dFOXO does not bind to the promoter of Atg8a ( Figure S6E ) . This is unlike mammalian FoxO3 which induces autophagy by directly binding to the promoters of LC3b , Gabarapl1 , and Atg12l in C2C12 myotubes [44] . Consistent with our model for negative regulation on Activin signaling by activated dFOXO , chico −/− inhibits Smox binding at Atg8a ( Figure 6D ) . An electrophoretic mobility shift assay ( EMSA ) confirms that Smox binds directly within Atg8a promoter . We expressed and purified a recombinant protein of the Smox-MH1 DNA binding domain ( amino acids 1–140 ) and measured its interaction with biotin-labeled Atg8a oligonucleotide probes containing Smad box ( 5′-AGAC AGAC-3′ ) . Smox-MH1 strongly bound to the Atg8a probe , and this interaction was blocked by addition of unlabeled wildtype cold probes ( Figure 6E ) . To define the required sequences of this Smad box ( AGAC ) we competed labeled wildtype probe with mutated cold probes . Unlabeled cold probes with mutations in both Smad boxes ( Mut2 and Mut3 ) did not compete with the wildtype binding , but cold probe mutants for only a single only Smad box ( Mut1 and Mut4 ) retained some competitive ability ( Figure 6E ) . Furthermore , in vitro expressed Smox-MH1 also binds to the oligonucleotide probe for the vertebrate Smad binding element ( 5′-GTCT AGAC-3′ ) ( Figure 6F ) . Together these data identify an invertebrate Smad binding element ( AGAC AGAC ) in the promoter region of the autophagy gene Atg8a . This Smad binding element contains a direct repeat of two Smad boxes ( AGAC ) . Upon activation , Smox , the Drosophila homologue of Smad2/3 , binds to the Smad box located within the promoter of Atg8a . Activin signaling represses autophagy via direct transcriptional regulation on the key autophagy gene Atg8a . To test whether increasing Atg8a expression within muscle is sufficient to promote lifespan , we over-expressing Atg8a using a muscle-specific driver ( MHC-Gal4 ) . Lifespan was modestly but significantly increased , suggesting that Atg8a gene is a specific instance of a longevity assurance genes that that functions through muscle downstream of Activin signaling ( Figure 7A , Table S4 ) . To further examine whether Atg8a is required for Activin-mediated lifespan extension , we silenced both Atg8a and daw using muscle-specific RNAi ( Figure 7B , Figure S7 , Table S5 ) . Lifespan extension when daw RNAi was expressed in muscle was rescued when Atg8a was simultaneously reduced by RNAi in this tissue , indicating Activin regulates longevity through muscle Atg8a . Previous studies from C . elegans recognize cross-talk between insulin/IGF-1 and TGF-β pathways [36] , [45] . DAF-16 is nuclear localized in many mutants of the TGF-β/dauer pathway [36] . To determine if muscle Activin signaling affects Drosophila lifespan through systemic insulin/IGF-1signaling we measured circulating insulin-like peptides in adults with tissue specific daw-RNAi . Knockdown of daw in muscle reduced the level of hemolymph DILP2 ( Figure 7C ) , while dilp2 mRNA remained constant ( Figure 7D ) , suggesting the daw specifically modulates DILP2 secretion . In contrast , knockdown of daw in fat body increased the level of circulating DILP2 ( Figure S6F ) . These contrasting tissue associated changes correspond to the observed effects upon lifespan when daw is reduced in each tissue ( Figure 3C–3H ) . Notably , dilp2 mRNA is reduced in other tissue-limited genetic manipulations that extend lifespan [19] , [46] , and knockout of the dilp2 locus is sufficient to extend lifespan [47] . We now see that reducing muscle Activin signaling via daw RNAi also remotely controls DILP2 secretion from the brain . This is sufficient to decrease systemic insulin signaling because insulin/FOXO sensitive 4ebp mRNA is elevated in peripheral tissues ( e . g . fat body ) ( Figure 7E ) . Unlike the consistent response of 4eBP in longevity mutants with reduced insulin , some but not all insulin/IGF pathway mutants have reduced fecundity . Here we found normal fecundity in females with reduced muscle Activin signaling ( Figure 7F ) , suggesting the effects of muscle Activin on lifespan are not mediated through trade-off between longevity and reproduction .
Insulin/IGF-1 signaling modulates longevity in many animals . Genetic analysis in C . elegans and Drosophila shows that insulin/IGF-1 signaling requires the DAF-16/FOXO transcription factor to extend lifespan , while in humans several polymorphisms of FoxO3A are associated with exceptional longevity [15] , [16] . Although many downstream effectors of FOXO have been identified through genome-wide studies [22] , [24] , [25] , [26] , the targets of FOXO responsible for longevity assurance upon reduced insulin signaling are largely unknown [24] . Here we found 273 genes targeted by Drosophila FOXO using ChIP-Seq with two long-lived insulin mutant genotypes . We focused on daw , an Activin ligand , which is transcriptionally repressed by FOXO upon reduced insulin/IGF signaling . Inactivation of daw and of its downstream signaling partners babo and Smox extend lifespan . These results are reminiscent of observations from C . elegans where reduced TGF-β/dauer signaling extends longevity [36] . Notably , the lifespan extension of TGF-β/dauer mutants ( e . g . daf-7 ( e1372 ) mutants ) can be suppressed by daf-16 mutants , suggesting that TGF-β signaling intersects with the insulin/IGF-1 pathway for longevity in C . elegans [36] . In our phylogenetic analysis , DAF-7 , Daw and mammalian Activin-like proteins share common ancestry . Activin signaling , in response to insulin/IGF-1 , may thus represent a taxonomically conserved longevity assurance pathway . Longevity benefits of reduced Activin ( TGF-β/dauer ) in C . elegans were resolved only when the matricide or ‘bagging’ ( due to progeny hatching within the mother ) was prevented by treating daf-7 ( e1372 ) mutants with 5-fluorodeoxyuridine ( FUdR ) to block progeny development [36] . This approach made it possible to distinguish the role of Activin in somatic aging from the previously recognized influence of BMP ( Sma/Mab signaling ) upon reproductive aging in C . elegans [35] , [48] . Activin , of course , is a somatically expressed regulatory hormone of mammalian menstrual cycles that induces follicle-stimulating hormone ( FSH ) in the pituitary gland . In young females , FSH is suppressed within a cycle when maturing follicles secrete the related TGF-β hormone Inhibin [49] . In mammalian reproductive aging , the effect of Activin in the pituitary becomes unopposed as the stock of primary follicles declines , thus inducing elevated production of FSH . We now find that reduced Activin but not BMP signaling favors somatic persistence in Drosophila . These parallels between reproductive and somatic aging among invertebrate models and humans suggest that unopposed Activin signaling is pro-aging while favoring reproduction . Reduced insulin/IGF signaling extends lifespan through interacting autonomous and non-autonomous actions . Reducing IIS in some distal tissues has been shown to slow aging because this reduces insulin secretion from a few neurons: reducing IIS by increasing dFOXO in fat body or muscle extends Drosophila fly lifespan while decreasing IPC production of systemically secreted DILP2 [18] , [19] . Here we identify Activin as a direct , downstream target of insulin/dFOXO signaling within muscles that has the capacity to non-autonomously regulate lifespan . Knockdown of Activin in muscle but not in fat body is sufficient to prolong lifespan . RNAi for muscle Activin signaling led to decreased circulating DILP2 and increased peripheral insulin signaling . Muscle is thus proposed to produce a signaling factor , a myokine , which impacts organism-wide aging and metabolism [18] , [50] , [51] ( Figure S8 ) . Aging muscle may produce different myokine-like signals in response to their physiological state . Aged muscles degenerate in many ways including changes in composition , mitochondria , regenerative potential and within-cell protein homeostasis [52] . Protein homeostasis is normally maintained , at least in part , by autophagy [40] , [53] . Loss of macroautophagy and chaperone-mediated autophagy with age will accelerate the accumulation of damaged proteins [54] . Expression of Atg8a in Drosophila CNS is reported to extend lifespan by 56% [28] , while recent studies find elevated autophagy in long-lived mutants including those of the insulin/IGF-1 signaling pathway [18] , [55] , [56] . Our results now show that insulin/IGF signaling can regulate autophagy through its control of Activin via dFOXO . Poly-ubiquitinated proteins accumulate in aging Drosophila while lysosome activity and macroautophagy decline . Muscle performance with age ( flight , climbing ) was preserved by inactivating Activin within this tissue . This genetic treatment also reduced the accumulation of protein aggregates . These effects are mediated by blocking the transcription factor Smox , which otherwise represses Atg8a . Smox directly regulates Atg8a through its conserved Smad binding motif ( AGAC AGAC ) . These results , however , contrast with an observation where TGF-β1 promotes autophagy in mouse mesangial cells [57] . Insulin/IGF-1 signaling is a widely conserved longevity assurance pathway . Our data indicate that reduced insulin/IGF-1 retards aging at least in part through its FOXO-mediated control of Activin . Furthermore , affecting Activin only in muscle is sufficient to slow its functional decline as well as to extend lifespan . Autophagy within aging muscle controls these outcomes , and we now find that Activin directly regulates autophagy through Smox-mediated repression of Atg8a . If extrapolated to mammals , pharmaceutical manipulations of Activin may reduce age-dependent muscle pathology associated with impaired autophagy , and potentially increase healthy and total lifespan through beneficial signaling derived from such preserved tissue .
Flies were reared and maintained at 25°C , 40% relative humidity and 12-hour light/dark . Adults were maintained upon agar-based diet with cornmeal ( 0 . 8% ) , sugar ( 10% ) , and yeast ( 8% unless otherwise noted ) . RU486 ( mifepristone , Sigma , St . Louis , MO , USA ) used to activate GeneSwitch-Gal4 was dissolved in ethanol to a concentration of 200 µM and added to the food . Fly stocks included: MHC-Gal4 [18] , Tub-GS-Gal4 [58]; da-GS-Gal4 [59]; S106-GS-Gal4 [58]; dilp2-GS-Gal4 ( provided by H . Jasper ) ; dilp2-GS-Gal4 , UAS-dicer2 ( provided by S . Helfand ) . RNAi lines for dFOXO target genes and TGF-β pathway are from Bloomington stock center ( TRiP line ) and Vienna Drosophila RNAi Center ( see the supplemental Table for the stock number ) . UAS-lines are: UAS-hairy [60] , UAS-puc [61] , UAS-RhoGAP18B [62] , UAS-vri [63] , UAS-wit [64] , UAS-cv-2 [65] , UAS-babo-Act ( also known as UAS-babo*1A2 ) [66] , UAS-Atg8-Cherry [67] . MHC-Gal4 is a constitutive muscle-specific driver , with expression restricted to the thoracic region and legs ( data not shown ) . Chico mutants were made in our lab as describe previously [13] , [29]: y1; cn1; ry506 ( wildtype ) , y1; cn1 chico1/cn1; ry506 ( chico−/+ ) , y1; cn1 chico1/cn1 chico1; ry506 ( chico−/− ) , y1; cn1 chico1/cn1 chico1; foxo21 ry506 ( chico−/− , foxo−/− ) . Adult on-set IPC ablation flies were made by crossing Dilp2-GS-Gal4 to UAS-rpr and inducing the cell death in IPC cells by feeding flies with RU486 for 15 days . Two insulin mutants were used in ChIP-Seq experiments , chico −/+ and IPC ablation . Chromatin immunoprecipitation ( ChIP ) was performed according to previously published methods with modification [68] , [69] , [70] . About 200–250 adult females ( ∼200 mg ) at the age of 15-day-old were pooled for each ChIP sample . Two biological replicates were prepared for each genotype . Flies were homogenized and cross-linked in 1× PBS containing 1% formaldehyde . The fly lysate were sonicated using a Branson 450 sonicator to break down the chromatin into a pool of DNA fragment with average size of 500 bp . Immunoprecipitation was performed using Dynal protean A beads ( Invitrogen , Grand Island , NY , USA ) and affinity purified anti-dFOXO antibody made in our laboratory . Following the wash with LiCl and TE buffer , the DNA-protein complex was eluted from the Dynal beads and reverse cross-linked . After Proteinase K digestion , dFOXO-bound DNA fragments were purified and diluted in Tris-HCl buffer . About 20 ng of ChIP DNA ( dFOXO-bound DNA ) and input DNA ( DNA sample before the immunoprecipitation ) were used in library preparation following the methods described in [71] . The libraries were then size-selected ( 150 bp-350 bp ) and purified by agarose gel , and subjected to the Illumina Genome Analyzer IIx Sequencer ( Illumina , San Diego , CA , USA ) . To map the dFOXO binding sites , we pooled the raw reads ( about 20 million reads per sample ) from two replicates into one data file and aligned it to Drosophila reference genome using Bowtie short read aligner [72] . About 70% of raw reads have at least one alignment . The enrichment of dFOXO binding between ChIP DNA and input DNA was determined using peak calling package PeakSeq [73] . Enriched regions with FDR of 0 . 01 were selected . Target genes , which were detected 5 kb away from the center of the binding sites , were also obtained . The ChIP-Seq raw data are archived at NCBI GEO with Accession # GSE44686 . For ChIP-PCR analysis , the binding enrichment was calculated as the fold change of ChIP DNA versus input DNA . The binding to the coding region of Actin gene ( Act5C ) and sry genomic region were used as negative controls . The DAVID functional classification tool was used for pathway and molecular function analysis on the dFOXO target genes [74] . Genomic sequence near the dFOXO binding region ( ∼200 bp ) was downloaded from the Flybase ( http://flybase . org/ ) and de novo motif analysis was performed using MEME Suite [75] . Total RNA was extracted from 10 whole flies or from tissue of 15 flies in Trizol reagent ( Invitrogen , Grand Island , NY , USA ) . DNase-treated total RNA was quantified with a NanoDrop ND-1000 . About 50–100 ng of total RNA was used for quantification with SuperScript One-Step RT-PCR reagent ( Invitrogen , Grand Island , NY , USA ) and measured on an ABI prism 7300 Sequence Detection System ( Applied Biosystems , Carlsbad , CA , USA ) . Three biological replicates were used for each experimental treatment . mRNA abundance of each gene was normalized relative to ribosomal protein L32 ( RpL32 , also known as rp49 ) by the method of comparative CT . Primer sequences are shown in Table S8 . Full length TGF-β ligands from worm , fly and mouse were aligned in ClustalW . From the alignments , a phylogenetic tree was constructed using MEGA 5 . 0 [76] , according to the neighbor-joining method with a bootstrap test calculated with 2000 replicates and a poisson correction model . Mouse Glial cell line-derived neurotrophic factor ( GDNF ) was used as the out-group . Two to three-day-old female adult flies were collected with light CO2 anesthesia and pooled in 1 L demography cages at a density of 100 to 125 flies per cages . Three independent cages were initiated per genotype . Food vials with media containing vehicle only or RU486 were changed every two days , at which time dead flies were removed and recorded . Survival analysis was conducted with JMP statistical software with data from replicate cages combined . Survival distributions were compared by the Log-Rank test . Cox proportional hazard survival analysis was used to assess how reduced daw and Atg8a interacted to affect mortality . Flying and climbing assays were scored as described in [18] . In the flying assay , flies were released at the top of a 250 ml cylinder ( about 30 cm long ) . The number of flies that didn't fall straight to the bottom of the cylinder was recorded . A total of 40 females were scored for each genotype . In the climbing assay ( also known as negative geotaxis assay ) , flies were first tapped down to the bottom of a standard ( empty ) food vial , and the percentage of flies that climbed up 8 cm within 20 seconds was recorded . A total of 80 females ( 10 flies per vial ) were scored for each genotype . Antibodies for immunostaining included: anti-polyubiquitin FK2 ( 1∶200 ) ( Assay Designs/Enzo Life Sciences , Farmingdale , NY , USA ) , and anti-rabbit IgG-DyLight 488 ( 1∶300 ) ( Jackson ImmunoResearch , West Grove , PA , USA ) . F-actin was visualized by Alexa Fluor 488-conjugated Phalloidin ( Invitrogen , Grand Island , NY , USA ) . Lysosome was monitored by LysoTracker Red DND-99 at the concentration of 100 nM ( Invitrogen , Grand Island , NY , USA ) . DNA was stained with Hoechst 33342 ( 1 µg/ml ) ( Invitrogen , Grand Island , NY , USA ) . Samples were processed as described in [18] , and imaged with a Leica SP2 laser scanning confocal microscope . To quantify the area of protein aggregates and the number of lysotracker or Atg8a-positive dots , grayscale images were converted to binary images ( halftone or black & white ) with a grayscale cutoff of 20 pixels using ImageJ software [77] . The number/area of positive immunostaining was measured with the “Analyze Particles” function . Smox polyclonal antibody was generated against the peptide sequence ( DSIVDYPLDNHTHQ ) corresponding to amino acids 143–156 ( Covance , Dedham , MA , USA ) and affinity purified ( Thermo/Pierce , Waltham , MA , USA ) ( specificity documented in Figure S5 ) . Phospho-Smad2 antibody was from Cell Signaling Technology ( #3108 ) ( Danvers , MA , USA ) . Thorax tissue from ten female adults was homogenized in RIPA buffer ( Thermo/Pierce , Waltham , MA , USA ) with protease inhibitor cocktail ( Sigma , St . Louis , MO , USA ) . Supernatant was incubated with SDS loading buffer ( Invitrogen , Grand Island , NY , USA ) at 70°C for 10 min . About 30 µg of denatured protein was separated on 10% SDS-polyacrylamide precast gels ( Invitrogen Grand Island , NY , USA ) and transferred to nitrocellulose membranes . Following incubation with primary and secondary antibodies , the blots were visualized with Pierce ECL Western Blotting Substrate ( Thermo Fisher Scientific , Waltham , MA , USA ) . Band intensity was quantified with Image Lab software ( Bio-Rad , Hercules , CA , USA ) . cDNA for Smox-MH1 DNA binding domain ( 1–420 nt ) was cloned into pFN29K-His6HaloTag protein expression vector ( Promega , Madison , WI , USA ) . After expression in E . coli , recombinant proteins were purified using HaloTag purification kit ( Promega , Madison , WI , USA ) . Empty vector was used as a negative control . Biotin-labeled DNA probes were generated using 3′-end biotin labeling ( Fisher/Thermo , Waltham , MA , USA ) . The binding reactions were carried out in a 10 µl of assay mixture containing 10 mM Tris·HCl ( pH 7 . 5 ) , 150 mM KCl , 5 mM MgCl2 , 10 ng/µL poly ( dI-dC ) , ∼50 ng labeled probe and 20 µg purified recombinant protein . After incubation at room temperature for 20 min , the mixtures were electrophoresed on 0 . 8% agarose gels in 0 . 5× Tris/borate/EDTA buffer . Biotin-labeled DNAs were transferred to a positive-charged nylon membrane ( Invitrogen , Grand Island , NY , USA ) and detected using GelShift Chemiluminescent EMSA ( Active Motif , Carlsbad , CA , USA ) . 10-day-old mated female flies were maintained on standard food ( 2% yeast ) for five days at three females per vial and 8–10 vials per group . Flies were passed daily to new vials over five days and eggs were counted daily . We followed our recently reported EIA assay to measure hemolymph DILP2 [21] . Briefly , about 0 . 5 µL of hemolymph was collected by decapitation of 15 female flies . Hemolymph was then incubated overnight in a 96-well EIA/RIA plate ( Corning Incorporated , Corning , NY , USA ) at room temperature . Anti-DILP2 antibody ( gift from P . Leopold ) was used at 1∶2500 dilution . After the incubation with a HRP-conjugated secondary antibody ( 1∶2500 ) , hemolymph samples were treated with TMB solution ( 3 , 3′ , 5 , 5′-teramethylbenzidine; American Qualex antibodies , San Clemente , CA ) ;absorbance was recorded at 450 nm upon a plate reader . Data are presented as mean ± SEM from three independent biological replicates , unless otherwise noted . Statistical significances were evaluated by t-test and one-way ANOVA analyses using GraphPad Prism Software . | It is widely known that reduced insulin/IGF signaling slows aging in many contexts . This process requires the forkhead transcription factor ( FOXO ) . FOXO modulates the expression of many genes , and the list of those associated with slow aging is impressive . But there are few data indicating the mechanisms or genes through which FOXO actually slows aging . Here , we identify a novel FOXO target , dawdle , the Activin-like ligand in fruit flies . We show that down-regulation of Activin signaling in muscle , but not in adipose tissue , leads to extended lifespan . In part it does so when it alleviates the negative transcriptional repression of its Smox transcription factor ( a Smad transcription factor ) upon a keystone autophagy gene , Atg8a . This double signaling cascade autonomously improves muscle performance ( measured at cellular and functional levels ) and nonautonomously extends lifespan as it reduces the secretion of insulin peptides from the brain . The work develops the emerging model for interacting autonomous-nonautonomous roles of insulin/IGF signaling as a systems integrative mechanism of aging control . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | Activin Signaling Targeted by Insulin/dFOXO Regulates Aging and Muscle Proteostasis in Drosophila |
For over two decades , a racemic mixture of oxamniquine ( OXA ) was administered to patients infected by Schistosoma mansoni , but whether one or both enantiomers exert antischistosomal activity was unknown . Recently , a ~30 kDa S . mansoni sulfotransferase ( SmSULT ) was identified as the target of OXA action . Here , we separate the OXA enantiomers using chromatographic methods and assign their optical activities as dextrorotary [ ( + ) -OXA] or levorotary [ ( - ) -OXA] . Crystal structures of the parasite enzyme in complex with optically pure ( + ) -OXA and ( - ) -OXA ) reveal their absolute configurations as S- and R- , respectively . When tested in vitro , S-OXA demonstrated the bulk of schistosomicidal activity , while R-OXA had antischistosomal effects when present at relatively high concentrations . Crystal structures R-OXA•SmSULT and S-OXA•SmSULT complexes reveal similarities in the modes of OXA binding , but only the S-OXA enantiomer is observed in the structure of the enzyme exposed to racemic OXA . Together the data suggest the higher schistosomicidal activity of S-OXA is correlated with its ability to outcompete R-OXA binding the sulfotransferase active site . These findings have important implications for the design , syntheses , and dosing of new OXA-based antischistosomal compounds .
For more than 25 years , the mainstay of treatment for Schistosoma mansoni infections in Brazil was the drug oxamniquine ( OXA , ( RS ) -1 , 2 , 3 , 4-tetrahydro-2- isopropylaminomethyl-7-nitro-6-quinolylmethanol ) [1 , 2] . OXA is species-specific , killing S . mansoni ( 67 million cases worldwide ) but not other schistosome species in Africa ( S . haematobium , 119 million cases ) or in SE Asia ( S . japonicum , 1 million cases ) [3 , 4] . OXA is no longer manufactured because the drug praziquantel , which is effective against all schistosome species , is now available at a reasonable price due to the expiration of its patent . The mode of action of OXA was recently elucidated[5] . As predicted by Pica-Mattoccia et al . [6] , OXA is a prodrug that is taken up by the parasite and sulfonated by an endogenous sulfotransferase ( SmSULT , GenBank AHB62207 . 1 , UniProt V9PWX8 ) in the presence of 3’phosphoadenosine 5’phosphosulfate ( PAPS ) . The resulting sulfate ester of OXA is an unstable species that spontaneously decays to form a reactive electrophilic product capable of alkylating DNA , proteins and other macromolecules . The ensuing disruption of synthetic and metabolic cellular functions eventually leads to parasite death[5 , 7] . OXA possesses one asymmetric carbon atom and its two enantiomers are both present in the marketed drug ( Fig 1 ) . The structure of SmSULT with bound OXA and depleted co-factor PAP was determined previously at a resolution of 1 . 75 Å [pdb code 4MUB , [5] . Although the crystals were soaked with racemic OXA , the structure revealed only S-OXA in the central cavity of the L-shaped , predominantly α-helical enzyme with its hydroxyl group ( the target of sulfonation ) centered at the end of a shaft running from the surface of the molecule . The relative positions of the accepting OXA and donating PAPS groups are entirely consistent with the formation of a sulfonated OXA hydroxyl group ( sulfate ester of OXA ) . Previous studies have succeeded in separating the enantiomers , but have not provided information about their absolute configurations or relative antischistosomal properties [8–10] . The question of which or if both enantiomers are active , and which or if both can occupy the binding pocket of SmSULT is addressed in this report .
Racemic OXA was a gift from Dr . D . Buggey ( Pfizer Ltd . ) . The separation of OXA enantiomers was carried out by HPLC on a modified cellulose chiral stationary phase ( Chiracell OD-H , Daicel Chemical Industry ) 4 . 6 mm i . d . x 150 mm , eluted at 1 mL/min with a mixture of 5% isopropanol in n-hexane ( HPLC grade , Carlo Erba , Italy ) , added with 0 . 1% dimethylamine ( Aldrich , 99 . 5% purified by distillation ) . The column consists of Cellulose tris ( 3 , 5-dimethylphenylcarbamate ) physically coated on microparticulate silica gel . This chiral selector is versatile and it shows a particularly good selectivity towards aromatic compounds with substituents containing N or O atoms . Dimethylamine was added to the eluent to prevent the ionization of OXA amino groups , but it was removed immediately after fraction collection , to avoid degradation of purified OXA isomers . The chromatogram was obtained at 254 nm . Fifty μL of a racemate solution ( 1 mg/mL ) in n-hexane:isopropanol ( 1:1 ) were injected into the column and the separated enantiomer peaks were collected . The racemate solution was stored at –20°C . In order to obtain sufficient material , the separation was repeated several times and the collected eluates were immediately evaporated to dryness under vacuum , in order to remove the dimethylamine . The pool of evaporated fractions corresponding to the two enantiomers were re-dissolved in 1 mL of n-hexane:isopropanol and 10 μL analyzed by HPLC in the chromatographic system described above . This allowed control of purity and quantitative determination by comparison with a calibration curve obtained with a known amount of racemate . A solution of 0 . 58 mg/mL of OXA racemate was prepared in CH3OH:H2O ( 60:40 ) and then diluted with the running buffer ( 50 mM pH 3 phosphate ) to obtain a 35 μg/mL solution . The CZE separation was performed as described by Abushoffa & Clark [8] . The background electrolyte consisted of running buffer with 1 mM heparin as a chiral selector . Separation was performed in a 62 cm , 75 μm i . d . capillary tube at 30°C , with an applied voltage of 20 kV . Samples were hydrodynamically injected ( 50 mbar , 5 sec ) . Fifty μL of enantiomer #1 ( i . e . the first eluting compound in HPLC ) was evaporated to dryness with a N2 flux and subsequently dissolved in 20 μL of 5% methanol in running buffer . Ten μL of this enantiomer #1 solution was added to 600 μL of racemate ( 35 μg/mL ) and the mixture was analyzed by CZE under the same conditions described above for the racemate . The sulfotransferase from S . mansoni ( SmSULT ) was crystallized as described previously [5] . The depleted co-factor PAP ( 3’phosphoadenosine 5’phosphate ) was added to achieve a 4 to 1 stoichiometric ratio over protein and incubated for one hour on ice prior to crystallization . Freshly grown crystals ( 4–8 days post-setup ) were soaked overnight in saturating conditions of each purified OXA enantiomer and flash-cooled in liquid nitrogen prior to data collection . All diffraction data were measured at the Advanced Photon Source NE-CAT beamline 24-ID-C and integrated and scaled using the program XDS [11] . Structures of the OXA enantiomer•SmSULT•PAP complexes were isomorphous with the published SmSULT structure ( Protein Data Bank entry 4MUA ) and the OXA enantiomers were built into difference electron density with coefficients mFo-DFc [12] . Model coordinates were refined using the PHENIX program suite [13] , including simulated annealing with torsion angle dynamics , alternating with manual model adjustment using the program COOT [14] . Figures depicting protein and OXA structure were created using the program PyMOL [15] . Coordinates and structure factors have been deposited in the Protein Data Bank [16] under accession codes 5BYJ and 5BYK . For the Institute of Cell Biology and Neurobiology , experimental protocols involving the use of animals were reviewed and approved by the Public Veterinary Health Department of the Italian Ministry of Health ( Authorization N . 25/2014-PR ) . For the University of Texas Health Science Center ( S2 Fig legend ) , the use of animals in his study was approved by the University of Texas Health Science Center IACUC ( Protocol 11087x ) that adheres to the NIH Animal Care and Use guidelines . A Puerto Rican strain of S . mansoni that has been maintained in the laboratory for several decades was used throughout this study . An albino strain of Biomphalaria glabrata served as the intermediate host , while CD1 female albino mice ( Harlan , Italy ) were used for the mammalian stages . Unisexual infections were obtained by exposing snails to a single miracidium and then sexing the emerging cercariae by PCR using female-specific W1 primers [17] . Mice infected by tail immersion with 160 male cercariae were perfused ≥7 weeks later and the worms obtained were used for drug assays . Eight to 13 male worms were distributed in tissue culture dishes ( 3 . 5 cm ) in Dulbecco modified Minimum Eagle’s Medium ( bicarbonate buffered ) supplemented with 10% fetal calf serum , 100 U/mL penicillin , 100 μg/mL streptomycin and 0 . 5 μg/mL amphotericin B . Cultures were kept at 37°C in an atmosphere of 5% CO2 in air and were observed daily under a Leica MZ12 . 5 stereomicroscope . Male worms were used as they are more sensitive to the effects of oxamniquine then are female worms [18] . Parasites were exposed to racemic OXA or its purified enantiomers for 30 min and subsequently washed three times and transferred to new dishes containing drug-free medium . At the end of the observation period ( 2 weeks at high doses; 3 weeks at low doses ) , worms were classified on the basis of various vitality indicators , as: normal ( similar to untreated controls ) and assigned score 100; slow ( decreased motility and slight morphological changes ) , score 60; moribund ( only tiny movements , marked morphological changes , opaque appearance ) , score 30; dead ( no movement , severe morphological changes , dark appearance ) , score 0 . The number of worms in each category was recorded . The scores of all worms were added , divided by the number of worms present in the dish and reported as average scores .
The enantiomers of OXA were separated by semi-preparative HPLC on a chiral column and the results are illustrated in Fig 2A . Two major peaks were present , practically with baseline separation , and were provisionally labeled as #1 and #2 . The area under the curve was essentially the same for the two peaks , consistent with the enantiomers being present in equal amounts . Since the quantity of compounds obtained from a single separation was limited , fractions #1 and #2 from several runs were pooled , respectively , and re-applied to the same column to check purity and to estimate quantity . As shown in Fig 2B and 2C , the separated enantiomers were reasonably pure and their total amounts were estimated to be about 200 μg for each enantiomer . The small amount of material required by CZE prompted us to use this technique in order to assign the optical activity of each enantiomer as either dextro- or levo-rotatory . In a previous separation by CZE , Abushoffa & Clark [8] showed that the OXA levorotatory ( – ) enantiomer has a higher electrophoretic mobility than the dextrorotatory ( + ) one . A racemate solution spiked with enantiomer #1 , obtained from the chromatographic separation , was then analyzed by CZE as previously described . Since compound #1 co-migrated with the faster peak of the racemic mixture , it was identified with the levorotatory ( – ) enantiomer ( Fig 3 ) . In our previous work , SmSULT crystals were soaked with racemic OXA prepared for medicinal use [5] . Although the preparation contained an approximate 1:1 mixture of enantiomers , only one ( S-OXA , see below ) was observed in the crystal structure . In the present study , single SmSULT crystals were soaked with purified preparations of either R- or S-OXA enantiomers . Data collection and refinement statistics for the two new structures determined in this study are shown in Table 1 . The OXA enantiomers bind in similar overall orientations ( Fig 4A and 4B ) . The crystal structures clearly reveal the chirality of each compound and identify the ( – ) -enantiomer ( peak #1 ) as R-OXA and the ( + ) -enantiomer ( peak #2 ) as S-OXA [19] . The positions and orientations of amino acid residues contacting the two enantiomers of OXA are virtually unchanged in the two structures and it is the relative positions of the two enantiomers of OXA and the water structure that adjust to best accommodate each enantiomer in the SmSULT binding cavity . Some of the ordered water structure in contact with OXA is preserved between enantiomer complexes including water molecules that form hydrogen bonds to the hydroxy moiety , the amine in the isopropylaminomethyl moiety , or make a van der Waals contact to the isopropyl moiety ( S1 Fig ) . A water molecule unique to the R-OXA complex is observed in hydrogen bonding distance to the piperidine amine . The orientation of S-OXA prevents a water molecule from occupying this same position , but two additional water molecules unique to the S-OXA complex are observed nearby within van der Waals contact distances ( S1 Fig ) . The isopropylaminomethyl and piperidine moieties ( Figs 1 and 5A and 5B ) of the OXA enantiomers are observed in different configurations . The terminal methyl groups of the isopropyl moiety in each enantiomer orient in the same direction and the adjacent secondary amino groups are both observed in hydrogen bonding distance to Asp144 ( Figs 4 and 5A ) . However , the chiral carbon atoms of the piperidine moieties joining the isopropylaminomethyl moieties in each enantiomer force the linked methyl groups to orient in opposite directions ( Fig 5B ) . Additionally , the piperidine rings bend in opposite directions at the carbon position next to the chiral carbon , relative to the plane of the rings thereby adopting opposite ring puckers ( Fig 5B ) . While the OXA nitro and hydroxymethyl substituent groups are rotated with respect to each other about an axis perpendicular to the aromatic ring , the accepting hydroxyl groups in both enantiomers essentially overlap and are therefore both in position to hydrogen bond to the side chain of Asp91 ( Fig 5A ) . The nitro moiety of S-OXA maintains a hydrogen bond ( 2 . 7Å ) with Thr157 as previously observed in the SmSULT complex structure determined in the presence of the racemic mixture . The nitro moiety of R-OXA is significantly rotated relative to that in S-OXA and positioned oriented at a greater distance from Thr157 ( 3 . 5Å ) suggesting a comparatively weak hydrogen bond ( Fig 5A ) . Overall , the ring structure of R-OXA rotates ~10 degrees about an axis perpendicular to the plane of the ring relative to S-OXA . The separated OXA enantiomers , together with the racemic mixture , were tested in vitro for their relative schistosomicidal properties . In these experiments , the concentration of racemate was double that of the enantiomers , under the assumption that only one of the two enantiomers might be active , so that the racemate would contain only 50% of active substance . In order to take into account as much as possible the variety of effects exerted on schistosomes by the different compounds , we preferred to adopt a vitality scoring system rather than a simple live/dead classification . Four different experiments were performed at the Institute of Cell Biology and Neurobiology , two of them at low doses ( upper half of Table 2 ) and two at high doses ( lower half of Table 2 ) . The results of the two experiments at each dose range were pooled and averaged . In the set of experiments carried out at low drug concentration , OXA racemate at 8 μg/mL reduced worm vitality to 14 . 4% of controls , while S-OXA ( at half that concentration ) was less effective , decreasing vitality to 32 . 6% of controls . R-OXA , on the other hand , had only a very modest effect , lowering worm vitality to about 89% of controls . Thus , under these conditions , S-OXA had about 3X the activity of R-OXA . When 5X higher drug concentrations were used ( lower half of Table 2 ) , the effect on worms was obviously more pronounced for all compounds , and even R-OXA displayed a sizeable activity ( bringing worm vitality down to 22 . 2% of controls ) ( see S1–S4 Videos ) . In an independent set of experiments conducted at UTHSCSA , using a different schistosome strain and minor methodological variations , adult schistosome male worms were treated with 40 μg/mL of racemate OXA and 20 μg/mL of each enantiomer . Under these conditions , S-OXA clearly showed a much higher activity than R-OXA , thus confirming the above results ( S2 Fig ) . Taken together , these results suggest that the bulk of antischistosomal activity is exerted by S-OXA , but even R-OXA can have antischistosomal effects when present at high concentrations . This is confirmed by the fact that the racemate at 8 μg/mL is more effective than S-OXA enantiomer at half the concentration , possibly due to a contribution of R-OXA to the overall activity . In order to rule out the possibility that the separated enantiomers might induce some non-specific toxicity , we had preliminarily ascertained that the OXA-resistant schistosome strain HR [5] was completely unaffected by these substances .
The separation of the two OXA enantiomers had been previously described using either chromatographic or electrophoretic approaches [8–10] , but the relative contributions of each enantiomer to antischistosomal activity and their individual involvement in the molecular mechanism of action had not been addressed . Data presented here show that the S- ( + ) -enantiomer is responsible for the majority of the antischistosomal activity , while the R- ( – ) -enantiomer is capable of exerting a moderate activity that is best detected when present alone and at relatively high concentration . Indeed , the crystal structures reveal both enantiomers bind similarly in the central cavity of SmSULT , although when crystals are exposed to a racemic mixture of OXA electron density for only S-OXA is observed . The structures of both enantiomer complexes with SmSULT reveal that the positions of amino acid residues surrounding OXA in the central cavity do not vary significantly . Instead , it is the orientation and configuration of the OXA enantiomers that varies in order to occupy the central cavity . Although R- and S-OXA occupy much of the same space in the cavity , the positions of the piperidine moiety and its isopropylaminomethyl substituent demonstrate the most variation due to influence of the chiral carbon while preserving the positioning of the methylhydroxy group which is the target of modification by the sulfotransferase . The combined results of these molecular and biological analyses suggest that when schistosomes are exposed to the racemic OXA mixture , the activating enzyme SmSULT preferentially binds and sulfonates the S-OXA , which has an overall better steric fit for the central cavity of the protein . The piperidine rings in the enantiomers show opposite puckers adjacent to the chiral carbon , but the adjoining isopropylaminomethyl groups of R- and S-OXA occupy similar positions in the binding pocket ( Fig 3 ) . R- and S-OXA also maintain similar hydrogen bonding distances ( 2 . 8 and 2 . 6 Å , respectively ) between the isopropylamino group and the Asp144 side chain Oδ . However , a more favorable hydrogen bond is formed by Thr157 to the nitro group of S-OXA ( 2 . 7 Å ) compared to R-OXA ( 3 . 5 Å ) . Thus , the S-OXA enantiomer may out-compete R-OXA due to a more favorable energy of binding in the racemic mixture . Kinetic data , which we are attempting to generate , may help address the issue of of why S is better than R in terms of activity . As with many other drugs , it would be desirable that stereochemically homogeneous compounds be employed as antischistosomal agents . Our ongoing efforts to generate novel wide spectrum OXA derivatives will definitely take this option into account . | Schistosomes , parasites that cause the disease schistosomiasis in humans , are blood flukes that infect an estimated 200 million people in 76 countries . Control of schistosomiasis is currently based on repeated doses of the drug praziquantel ( PZQ ) . Parasites showing reduced susceptibility to PZQ have been recovered from patients that failed PZQ treatment and have been obtained by experimental selection . New anti-schistosomal drugs are therefore needed that can be used with PZQ to minimize the probability of resistance . The older anti-schistosomal drug oxamniquine ( OXA ) has an excellent efficacy and safety record but is only active against one of the three species infecting humans . Recently , a combination of genetic and structural analyses resulted in the determination of the structure of OXA in complex with its target enzyme in the parasite , providing opportunity for structure-guided modifications of OXA to make it more effective against all three schistosome species . Synthesis of OXA results in a racemic mixture . Here , we isolate OXA enantiomers and find that one is more effective than the other at killing schistosomes . Crystal structures of both OXA enantiomers bound to the target enzyme suggest a molecular basis for this observation that should be considered in ongoing and future OXA-based drug design efforts . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Structural and Functional Characterization of the Enantiomers of the Antischistosomal Drug Oxamniquine |
Telomerase activity is readily detectable in extracts from human hematopoietic stem and progenitor cells , but appears unable to maintain telomere length with proliferation in vitro and with age in vivo . We performed a detailed study of the telomere length by flow FISH analysis in leukocytes from 835 healthy individuals and 60 individuals with reduced telomerase activity . Healthy individuals showed a broad range in average telomere length in granulocytes and lymphocytes at any given age . The average telomere length declined with age at a rate that differed between age-specific breakpoints and between cell types . Gender differences between leukocyte telomere lengths were observed for all cell subsets studied; interestingly , this trend could already be detected at birth . Heterozygous carriers for mutations in either the telomerase reverse transcriptase ( hTERT ) or the telomerase RNA template ( hTERC ) gene displayed striking and comparable telomere length deficits . Further , non-carrier relatives of such heterozygous individuals had somewhat shorter leukocyte telomere lengths than expected; this difference was most profound for granulocytes . Failure to maintain telomere homeostasis as a result of partial telomerase deficiency is thought to trigger cell senescence or cell death , eventually causing tissue failure syndromes . Our data are consistent with these statements and suggest that the likelihood of similar processes occurring in normal individuals increases with age . Our work highlights the essential role of telomerase in the hematopoietic system and supports the notion that telomerase levels in hematopoietic cells , while limiting and unable to prevent overall telomere shortening , are nevertheless crucial to maintain telomere homeostasis with age .
At least a few hundred nucleotides of telomere repeats must “cap” each chromosome end in order to suppress DNA damage signals and avoid the activation of DNA repair pathways [1]–[3] . Critically short or “uncapped” telomeres may be repaired by the enzyme telomerase [4] or by recombination [5] . However , the capacity of these telomere repair processes appears limited in most human somatic cells [6] . Apoptosis or cellular senescence is triggered when too many “uncapped” telomeres accumulate [7] , posing a barrier to tumor growth , but also contributing to loss of cells with age [8] . Despite increasing evidence that telomere homeostasis is important in human aging , cancer and disease states , detailed and comparative information regarding the telomere length in different human cell subtypes of healthy individuals in relation to their age is surprisingly modest . Apart from being technically challenging [9] such studies are complicated because at birth and throughout life , telomere length is highly variable between chromosomes [10] , [11] , between cells [12] , [13] and between individuals . Studies of identical twins have shown that individual differences in average telomere length appear to be largely genetically determined [14] , [15] . In most somatic cells the telomere length declines with age and with cell division in culture , albeit at different rates [13] , [16] . For example , in humans and baboons , lymphocytes show a more pronounced telomere loss with age than granulocytes [14] , [17] . These two cell types represent the two major branches of the hematopoietic system , which can be further subdivided into distinct cell populations based on their phenotype and function . Within the hematopoietic hierarchy , the most primitive cells , hematopoietic stem cells ( HSC ) , have the longest telomeres [18] , [19] . HSC differentiate to produce progenitor cells of both the myeloid and lymphoid lineage that proliferate prior to differentiation into mature “end” cells . Unlike most immune cells most differentiated myeloid cells such as granulocytes are incapable of further cell divisions . The precise role of telomerase in hematopoietic stem and progenitor cells and in lymphocytes remains poorly understood . Telomerase expression is readily detected in hematopoietic cells [20]–[22]; however , this activity appears unable to prevent telomere loss with age or proliferation . It is often assumed that telomerase is required to maintain the telomere length in various stem cells . With the exception of embryonic stem cells and abnormal tumor ( stem ) cells this assumption is not supported by data . Studies on the role of telomeres and telomerase in HSC from healthy individuals are challenging because HSC are very rare cells that typically reside in bone marrow . In contrast , the various nucleated blood cells that are derived from HSC are easily accessible for study . The average telomere length in granulocytes can be used as a surrogate marker for the telomere length in HSC [23] , if one assumes that the number of cell divisions between HSC and granulocytes is relatively constant [18] . Individual carriers of heterozygous mutations for either the telomerase RNA gene ( hTERC ) or the telomerase reverse transcriptase gene ( hTERT ) can present with a wide spectrum of diseases [24] including dyskeratosis congenita [25] , [26] , bone marrow failure [24] and pulmonary fibrosis [27] . Heritable telomerase deficiencies provide an excellent model to study the role of telomerase in human hematopoietic cells . Here we report our data on the median telomere length ( MTL ) in five distinct leukocyte subpopulations of over 800 healthy individuals between birth and 100 years of age as well as 60 individuals that are heterozygous for one of the telomerase genes , hTERC or hTERT . The telomere length in leukocytes from healthy individuals was found to vary over a broad range at any given age and the rate of telomere attrition also varied with age and with cell type . Strikingly , the telomeres in cells from individuals with telomerase deficiency were found to be very short in all cell types , and this deficit was found to be comparable for most cell subtypes for hTERC or hTERT deficiency . The largest ( age adjusted ) differences in telomere length deficits between hTERC or hTERT were seen in “naïve” T cells for hTERC deficient individuals and in NK/differentiated T cells for hTERT deficient individuals . These results demonstrate that normal telomerase levels are essential to maintain normal telomere homeostasis in HSC and lymphocytes . Our results provide valuable reference data for further studies of telomere biology in health and disease and point to a crucial rate-limiting role for telomerase in HSC and immune cells .
We measured the telomere length in lymphocytes and granulocytes of 835 healthy individuals using automated multicolor flow FISH ( Figure 1 ) . On average , 7 to 8 individuals were tested for each age-year . Various best-fit models were tested to model the overall decline in telomere length with age . In view of the very rapid decline in telomere length in the first years of life in humans [14] as well as non-human primates [28] , we divided the telomere length decline over three age segments . The first is between birth and one year of age when the growth rate of bones and weight in infants shows a marked deceleration ( for reference curves , see http://www . cdc . gov/growthcharts/clinical_charts . htm ) . A second arbitrary cut-off was set at 18 years of age because the decline in telomere length in all leukocytes appeared to drop notably after puberty . Telomere length data within the three selected age segments: below 1 year ( yr ) , 1–18 yrs and 19 yr and higher are shown in Figure 1 and Table 1 . The overall age-related telomere length decline was most pronounced in lymphocytes with significant losses ranging from 1190 base pairs ( bp ) per year between birth and 1 year of age to 126 bp per year during childhood and 43 bp per year in adulthood . In contrast , the age-related telomere length decline in granulocytes and by extension in HSC was more modest during early life ( 485 bp per year ) , childhood ( 74 bp per year ) and adulthood ( 28 bp per year ) . Telomere length measurements versus age in granulocytes and lymphocyte subpopulations were used to determine the regression lines for telomere attrition in the three selected age ranges ( regression estimates shown in Figure 2A; the complete data set can be accessed in Table S1 ) . These regression lines were shifted according to data distribution ( from the overall regression estimate ) to represent the 99th , 90th , 10th and 1st percentile of the telomere length distribution in each age segment for each blood cell subset in healthy individuals . The rate of telomere length decline varied amongst the different lymphocyte subsets analyzed . The telomere length decline with age in B lymphocyte subset ( CD45RA+ CD20+ ) was comparable to that in granulocytes . Memory T ( CD45RA−CD20− ) and mature NK/T ( CD45RA+CD57+ ) lymphocyte subsets showed the sharpest decline in telomere length with age , particularly during childhood with slopes of −144 and −155 bp per year respectively . The CD45RA+CD20− T lymphocyte subset enriched for “naïve” T cells and the CD45RA+CD57+ mature NK/T lymphocyte subset displayed the widest distributions , 2 . 80 and 2 . 89 kilobase ( kb ) respectively between the 10th and 90th percentile of the normal distribution , throughout the age ranges . Unlike other subsets CD45RA+CD20− T lymphocytes showed only a modest difference in the telomere attrition rate between childhood and adulthood: 89 and 51 bp per year respectively . In contrast , the memory T lymphocyte subset ( CD45RA−CD20− ) displayed the narrowest range of telomere length distribution ( 2 . 28 kb between the 10th and 90th percentile of the normal distribution ) . Overall , the shortest telomere lengths were measured in memory T and mature NK/T lymphocytes from older individuals . From our cross-sectional data , we determined the average telomere length decline with age for the different leukocyte subpopulations ( Figure 2B ) . During childhood , granulocytes , CD45RA+CD20− “naïve” T lymphocytes and CD20+ B lymphocytes all showed a very similar decline in telomere length , whereas the rate of decline in memory T cells was much higher . Paired MTL values in different blood cell subsets from the same individual revealed that around one year of age the telomere length values in memory T lymphocytes drop below those of granulocytes ( Figure 2C ) . In contrast , telomere length values in B lymphocytes remained comparable to those in granulocytes over the entire age range . One caveat in our measurement of telomere length in “naïve” ( CD45RA+CD20− ) T lymphocytes is that terminally differentiated effector lymphocytes re-expressing CD45RA are likely to represent an increased proportion within this cell population in older individuals . As a consequence , measurements within the subset of “naïve” T cells are variably skewed in older individuals ( as illustrated by the direct comparison of MTL between “naïve” T lymphocytes and other cell subtypes from the same individual over 4 distinct age groups in Figure S2A and Table S2 ) . Measurements from cord blood samples provided the earliest opportunity to assess the telomere lengths in cells from healthy individuals . Interestingly , of all the cell types measured from cord blood , granulocytes showed the shortest telomeres at birth: differences were significant for granulocytes versus CD45−CD20− memory T lymphocytes and CD45+CD20− “naïve” T lymphocytes but not for granulocytes versus B lymphocytes . Comparisons were tested by one-way ANOVA ( n = 58 ) : F ( 5 , 265 ) = 5 . 7; P = 0 . 0002 , Table S3 , followed by Tukey's multiple comparison test , see details in Table S4 ) . Interestingly , female newborns appeared to have longer telomeres than males ( Figure 3A ) ; however this trend did not reach statistical significance . Further comparisons of telomere length in leukocyte subsets as a function of gender showed highly significant differences between males and females in the CD45RA+CD20− “naïve” T lymphocyte subset over the entire age range ( F ( 4 , 825 ) = 9 . 05; P = 3 . 7×10−7 , ANOVA test result comparing regression fits; Figure 3B and Table S5 ) . Significant differences were also seen for other leukocyte subsets in each age segment with the exception of granulocytes and memory T lymphocytes , which displayed similar , average telomere lengths after 18 years of age ( Figure S3 and Table S5 ) . To study the role of telomerase in hematopoietic cells , we analyzed the telomere length in leukocyte subpopulations of individuals carrying a mutation in either hTERT or hTERC ( n = 60 ) in comparison to non-carrier relatives ( n = 37 ) . The results , plotted on the telomere length versus age distribution curves derived from healthy individuals ( Figure 2A ) are shown in Figure 4 and Table 1 . Strikingly , telomerase heterozygous individuals showed very short telomeres ( typically below the 1st percentile of the normal distribution ) at all ages and for all blood cell subsets tested ( ANOVA test P<2 . 2×10−16 , for full details of analyses see Table S5 ) . The shortest telomeres were measured in mature NK/T cells ( mean of 3 . 7±0 . 7 kb for all telomerase heterozygous individuals not adjusted for age , Figure 4A ) and the CD45RA+CD20− “naive” lymphocyte subset appeared the most severely impacted by telomerase deficiency with an average difference to the normal distribution ( adjusted for age ) of Δtel: 3 . 2 kb . Differential analysis of leukocyte telomere lengths in leukocytes from hTERT vs . hTERC heterozygous individuals showed a similar effect on most blood cell subtypes ( Figure 4B and Table 2 ) . Exceptions were an increased telomere loss in the CD45RA+CD57+ mature NK/T cells ( difference of 0 . 4 Kb ) of hTERT deficient individuals ( n = 37 ) and a slightly increased effect ( difference of 0 . 2 Kb ) on the CD45RA+CD20− “naive” lymphocyte subset for hTERC deficient individuals ( n = 23 ) . Non-carrier relatives of telomerase deficient individuals ( hTERT and hTERC considered together ) , despite having intact telomerase genes , also showed somewhat shorter median telomere lengths in all leukocytes compared to the control population . The largest difference was measured in the granulocyte subset of non-carrier relatives considered together , with Δtel: 0 . 9 kb , which may be indicative of HSC deficit and warrants further investigation ( Figure 4A; ANOVA test: F ( 3 , 837 ) = 8 . 1; P = 2 . 63×10−5 , for full details of analyses see Table S5 ) . Both parents and siblings of heterozygous individuals were found to have slightly shorter telomere lengths for age ( Table 2 and Figure S4 ) . Differential analysis of parents ( n = 6 ) and siblings ( n = 4 ) of hTERT deficient individuals was also performed and showed comparable telomere length deficits for all cell subsets tested ( Figure S4 and data not shown ) in this relatively small group .
In this report , we show telomere length data for five distinct leukocyte subpopulations from over 800 healthy individuals , representing a comprehensive and representative cross-sectional analysis of telomere length in leukocyte subpopulations over the entire human life span . The value of this data is illustrated by our analysis of individuals with heritable telomerase deficiencies . Leukocyte telomere length was found to clearly distinguish between relatives with and without mutations in hTERT or hTERC supporting telomere length measurements as a screen for mutations in “telomere maintenance” genes . Our results confirm and extend earlier reports of telomere loss in leukocytes with age [13] , [14] and document a crucial role for telomerase in controlling leukocyte telomere length . Telomerase expression is readily detected in most hematopoietic cells [20]–[22] , yet this activity appears unable to prevent the overall loss of telomeric DNA with age or proliferation . Most likely , telomerase is primarily required to directly act on chromosome ends in hematopoietic cells themselves , however secondary , indirect effects of telomerase via cells that support cell proliferation [29] or possible effects of the TERT protein on transcription in stem cells [30] are difficult to exclude . Heterozygosity for one of the telomerase genes , expected to reduce telomerase levels by half , results in a striking telomere deficit ( Figure 4 ) . How can this finding be explained ? One possibility is that the primary function of telomerase in somatic cells is the repair [8] or protection [31] of critically short telomeres . Failure to properly “cap” all chromosome ends with telomere repeats results in activation of a DNA damage response [1] , [32] . Detrimental consequences for HSCs and lymphocytes could result when DNA damage signals from uncapped telomeres persist or reach a certain threshold and cause apoptosis of such cells . Impaired “capping” of telomeres in cells with reduced telomerase could affect telomere length directly and indirectly . Direct effects on telomere length could result from normal replication of telomeric DNA [8] and damage caused by reactive oxygen species [33]–[35] . Indirect effect on telomere length would result from the additional cell division required to compensate for the increased cell losses . Compensatory cell divisions in cells from telomerase deficient individuals could be particularly taxing as more short telomeres are expected to emerge with each extra cell division . The resulting feed-forward loop could exhaust the stem cell compartment in infants and children explaining the marrow failure typically seen in pediatric telomerase deficient patients . In cases where sufficient stem cells survived till adulthood , the same unproductive feed-forward loop could exhaust cells of the immune system . This possibility is in line with the observation that after puberty telomere attrition in more mature subsets of T and NK cells is notably higher than in granulocytes as a surrogate marker for stem cells ( Figure 2 ) . We speculate that the balance between end cells such as granulocytes and macrophages on the one hand and various other immune cell types on the other is perturbed in older telomerase deficient patients . Such an imbalance could result in failure to clear pathogens and immunogens and create pro-fibrotic conditions or result in failure to remove senescent cells [36] . Apart from cell turnover and telomerase levels , the telomere length in parental chromosomes at fertilization is another probable variable in the disease manifestations of telomerase deficient patients . This variable will determine when critically short telomeres , requiring repair or capping by telomerase , will appear: during development or during adult life . This notion is in line with the age-related onset of symptoms or “anticipation” in multi-generation telomerase deficiency disorders [24] , [37] and our observation that telomeres in cells from unaffected children and parents of telomerase heterozygous individuals are somewhat shorter than expected ( Figure 4 and Table 2 ) . As was shown previously for human lymphocytes and granulocytes [14] , [38] and confirmed in longitudinal studies of non-human primates [28] , leukocyte telomere length shortens most dramatically very early in life . This rapid decline can be explained by steady proliferation of stem cells and immune cells after birth . After one year of age we observed a rapid deceleration in telomere loss most likely reflecting an intrinsic , ontogeny-related change in stem cell turnover and function [39] , which has also been observed in postnatal mice [40] , [41] . Our observations with human and primate cells suggest that each HSC cell division in these species is “counted” by the loss of telomeric DNA . Why a relative modest decline in telomerase activity in humans results in a wide spectrum of diseases whereas complete loss of telomerase is typically tolerated for several generations in yeast , plants , worms and mice remains incompletely understood [42] . The shortest overall telomere lengths were measured in the mature NK/T cell subsets of older healthy individuals and of hTERT telomerase-deficient individuals . These results suggest that one of the primary consequences of telomere attrition and telomerase deficiencies could be the loss of NK immune function . This notion is compatible with the reported age-related decline in the number and function of these cells [43] . Of note , in some individuals the estimated telomere length in mature NK/T cells was near the predicted minimal telomere length ( represented as a shaded area in Figure 1 and Figure 2 ) meaning that on average each chromosome end in those cells has fewer than 1 kb of telomere repeats . Despite the finding that at birth telomeres in lymphocytes are longer than in granulocytes and despite the selective expression of telomerase in cells of the lymphoid lineage upon activation [22] , [44] , T lymphocytes displayed a sharp decline in telomere length with age . The steady decline in the telomere length in T cells likely contributes to compromised adaptive immunity in the elderly [45] and in individuals with telomerase deficiencies [46] . Interestingly , the narrowest telomere length distribution in leukocyte subsets from healthy individuals was observed in the memory T cell compartment , pointing to a possible role of telomere length in shaping the T cell repertoire and immune memory . Our study of gender specific differences in telomere length confirmed previous observations that telomere lengths on average appear to be somewhat longer in females than in males [47] . The fact that this trend is already seen at birth raises questions that warrant further investigation: do females have fewer HSC at birth ? Do female HSC have a higher replicative potential because of longer telomeres ? Do stem cells in females have higher telomerase activity , possibly influenced by levels of sex hormones [48] or do other factors explain the longer telomeres in female leukocytes ? Epidemiological studies have been conducted to examine the potential validity of using relative leukocyte telomere length as a disease or aging associated biomarker . Interest in this area has greatly increased following recent reports of associations of shorter leukocyte telomere lengths with morbidity ( such as cardiovascular disease or diabetes reviewed in [49] ) and in response to external factors such as chronic stress [50] . More data is needed to confirm these findings and establish whether shorter leukocyte telomere lengths are associated with overall increased mortality in older adults [51] , [52] and whether the increased risk of infection such as pneumonia in elderly individuals differs significantly in relation to their telomere lengths . In conclusion , the data presented here contribute valuable base-line information regarding the telomere length in subpopulations of leukocytes during normal human ageing . This information will be a useful reference in studies of a variety of health conditions . Our data show that suppression of half of telomerase levels over a lifetime can severely compromise the telomere homeostasis of granulocytes as a surrogate marker for HSCs , and of immune cells . This likely is a dominant factor in the serious impairment of cell function and proliferative capacity that has been documented in telomerase deficient individuals . It seems possible that more effective short-term inhibition of telomerase could compromise the function of hematopoietic cells more acutely . Most likely , limitations imposed by progressive telomere loss act as a tumor suppressor mechanism in long-lived animals [42] . If so , caution is also needed for strategies that aim to rejuvenate older cells by reactivation of telomerase . The telomere length data described in this paper provide reference data for therapeutic strategies that target telomerase and for further studies on the role of telomeres and telomerase in normal aging and a variety of pathological conditions .
All subjects enrolled in this study in Vancouver signed informed consent forms that were approved by the University of British Columbia ( BC ) and BC Cancer Agency Research Ethics Board . All samples from patients outside Vancouver were obtained with informed consent and approval of local ethical review boards in accordance with the Declaration of Helsinki . Anonymous cord blood samples were obtained from healthy full term births with parental informed consent . Since no associate information was available for these samples , gender testing was performed by FISH as described below . Anonymous peripheral blood samples were obtained from 835 healthy individuals between the ages of 6 months to 102 years of age screened for clotting disorders; samples where no clotting disorders were found were made available for study; only gender and age information were provided . Samples from 60 individuals with confirmed telomerase deficiencies due to heterozygous mutations for either the telomerase reverse transcriptase ( hTERT ) or the RNA template ( hTERC ) gene and their 37 ( non-carrier ) relatives were included in our analysis and were described previously , ( mean ages for both groups were 41 and 45 years respectively [25] , [53]–[59]; all 97 participants or their guardians provided written informed consent in accordance with the Declaration of Helsinki . X and Y chromosome specific FISH was preformed as previously described [60] . Briefly , nucleated cord blood cells were fixed with methanol–acetic acid then dropped onto slides . Slides were fixed with formaldehyde , treated with pepsin , and dehydrated with ethanol . The hybridization mix containing fluorescently labeled peptide nucleic acid ( PNA ) probes specific for centromere repeats of respectively the X chromosome and the long arm of the Y chromosome were added to the slides . Following denaturation of DNA at 80°C for 3 minutes slides were incubated at room temperature for 30 minutes , washed , counterstained with DAPI and mounted using DABCO anti-fading reagent ( Sigma Aldrich ) . Images were acquired and analyzed as previously described [60] . hTERT and hTERC genotyping was performed as described previously [25] , [53]–[59] . Telomere length measurements using automated multicolor flow-fluorescence in situ hybridization ( flow FISH ) was performed as described [61] . Briefly , white blood cells ( WBCs ) were isolated by osmotic lysis of erythrocytes in whole blood using NH4Cl . The WBCs were then mixed with bovine thymocytes of known telomere length ( which serve as an internal control ) , denatured in formamide at 87°C , and hybridized with a fluorescein-conjugated ( CCCTAA ) 3 peptide nucleic acid ( PNA ) probe specific for telomere repeats and counterstained with LDS751 DNA dye . The fluorescence intensity in , granulocytes , total lymphocytes and lymphocyte subsets defined by labeled antibodies specific for CD20 , CD45RA and CD57 relative to internal control cells and unstained controls was measured on a FACSCalibur instrument ( Becton Dickinson ) to calculate the median telomere length from duplicate measurements . Further details regarding telomere length measurements and data sets are described in Online Supplementary Material as well as depicted in Figure S1 . Some of the total 835 healthy subjects did not have sufficient cells for analyses of one or more of the cell subsets tested: granulocytes , B lymphocytes , or mature NK/T cells . Specific improvements were developed during the 8 year of the healthy donor study allowing for the testing of additional cell subsets ( B and mature NK/T [62] ) explaining why fewer measurements are reported for these subsets . In addition , a slight modification was made to the cell lysis protocol: from a semi-automated small volume lysis in a 96 well format [63] ) to the current larger volume individual sample lysis [61] . The data obtained during these two experimental periods were first analyzed separately to test for differences between the first and second data sets . Briefly , both data sets were found to have comparable telomere length distributions over age , with a small but notable decrease in the calculated granulocyte telomere length together with a decrease in the range of granulocyte telomere length values in the second data set ( Figure S1 ) . These differences may be explained by the protocol improvements in the second set that resulted in a better resolution of signal and a reduction in the background fluorescence observed in cell types with large volumes of cytoplasm . Since the ranges lower limits were similar between the two data sets and since the majority of the data for both sets falls within the same 10th to 90th percentile range , the two sets were merged and analyzed together for curve fitting models and reference comparisons . Analyses were performed using Microsoft Excel ( Microsoft Office 2007 ) , GraphPad Prism ( version 4 ) and R ( version 2 . 6 . 1 , 2007 , The R Foundation for Statistical Computing ) ; t-Tests were two-tailed and performed on data with a normal distribution ( KS test ) . Linear modeling ( lm function in the R language ) was used to carry out the regression analysis and estimate the piecewise linear curves with breakpoints hinged at 1 and 18 years of age . We found that this gave the best fit with the least mean square error and more consistent error distribution across the age ranges as compared to using a number of polynomial fits ( linear , quadratic , cubic or quartic ) . The 99th , 90th , 10th and 1st percentile curves were obtained by vertical shift of the estimated regression curve to span the desired number of data points in the primary data set of healthy subjects ( n = 835 ) that were representative for this segment of distribution . To compare the telomere lengths of one population against another , the ANOVA function in the R language was employed to test if the data was from the same or 2 different model fits [64] . Figure S1 depicts data from two consecutive experimental periods separately for lymphocytes and granulocytes respectively , and displays the previous quadratic curve fitting model ( first ∼400 data points ) compared to the current three piece-wise linear regression model ( for which first and second data sets were combined ) used in Figure 2 , Figure 3 , and Figure 4 . Figure S2 complements Figure 2 and highlights the telomere length skewing specifically seen in CD45RA+ CD20− lymphocytes of older individuals , where a higher proportion of terminally differentiated lymphocytes are likely present within the cell population with this phenotype . Further , Figure S2 depicts the skewing observed in telomerase heterozygous individuals , comparable to that seen in healthy older individuals ( over 75 years of age ) . Figure S3 complements Figure 3 and depicts gender segregated telomere length data measured by flow FISH for all leukocyte cell subsets tested , together with the model fit statistical test results from these analyses . Figure S4 complements Figure 4 and depicts further analysis of leukocyte telomere length data from direct relatives , siblings or parents of telomerase heterozygous individuals . From this relatively small group , although a trend towards shorter MTLs is observed , no statistical difference between the groups and no statistically significant difference compared to healthy individuals was detected ( Table S5 ) . Table S1 displays the complete telomere length data sets . Duplicate leukocyte telomere length data were collected over an eight year period and analyzed on two occasions ( first analysis after 391 samples and the present analysis after next 445 samples ) . For the first set of data , freshly isolated nucleated blood cells were used whereas for the second set , nucleated cells were frozen prior to flow FISH ( see Materials and Methods and data set comparisons , Figure S1 ) . No marked differences in the calculated telomere length between the two data sets were observed for lymphocytes or lymphocyte subsets . Although the granulocyte telomere length values were slightly lower and more narrowly distributed in the second data set , the overall results were pooled for the current analysis . Table S2 displays telomere length cell subset correlations at different age ranges . This table complements Figure 2 and Figure S2 . It displays the correlative r values between paired cell population telomere length values . The cell population chosen as a reference has a set value of 1 . Tables S3 , S4 and S5 display the complete statistical ANOVA analysis results for comparing telomere lengths of leukocyte subsets in cord blood ( at birth ) , and comparing the two linear models ( an estimate of the entire population ) and an estimate of where factor “X” ( gender for example ) was taken into consideration and showed a significant difference . | Human blood cells all originate from a common precursor , the hematopoietic stem cell . Telomerase , the enzyme responsible for adding telomere repeats to chromosome ends , is active in human hematopoietic stem cells but appears unable to maintain a constant telomere length with age . We first document the telomere length of different blood cell subsets from 835 healthy individuals between birth and 100 years , to delineate the normal rate of telomere attrition with age . Telomere lengths of blood cells were found to be slightly longer in women than in men , from birth and throughout life . We then compared this reference data to the telomere length in similar blood cell subsets from individuals with reduced telomerase activity as a result of a mutation in one of the genes encoding telomerase and from their direct relatives . Strikingly short telomeres were found in telomerase-deficient individuals , consistent with their cellular pathology and disease susceptibility , and somewhat shorter telomeres than expected were found in cells of relatives with normal telomerase maintenance . Our data can be used as a reference for blood cell telomere length in studies of normal and accelerated aging . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"hematopoiesis",
"genetics",
"genetics",
"and",
"genomics",
"biology",
"genetics",
"of",
"disease",
"hematology"
] | 2012 | Collapse of Telomere Homeostasis in Hematopoietic Cells Caused by Heterozygous Mutations in Telomerase Genes |
Steroid hormones are crucial for many biological events in multicellular organisms . In insects , the principal steroid hormones are ecdysteroids , which play essential roles in regulating molting and metamorphosis . During larval and pupal development , ecdysteroids are synthesized in the prothoracic gland ( PG ) from dietary cholesterol via a series of hydroxylation and oxidation steps . The expression of all but one of the known ecdysteroid biosynthetic enzymes is restricted to the PG , but the transcriptional regulatory networks responsible for generating such exquisite tissue-specific regulation is only beginning to be elucidated . Here , we report identification and characterization of the C2H2-type zinc finger transcription factor Ouija board ( Ouib ) necessary for ecdysteroid production in the PG in the fruit fly Drosophila melanogaster . Expression of ouib is predominantly limited to the PG , and genetic null mutants of ouib result in larval developmental arrest that can be rescued by administrating an active ecdysteroid . Interestingly , ouib mutant animals exhibit a strong reduction in the expression of one ecdysteroid biosynthetic enzyme , spookier . Using a cell culture-based luciferase reporter assay , Ouib protein stimulates transcription of spok by binding to a specific ~15 bp response element in the spok PG enhancer element . Most remarkable , the developmental arrest phenotype of ouib mutants is rescued by over-expression of a functionally-equivalent paralog of spookier . These observations imply that the main biological function of Ouib is to specifically regulate spookier transcription during Drosophila development .
Steroid hormones are responsible for the coordination and regulation of many biological events during development of multicellular organisms . In all species , steroid hormones are synthesized from cholesterol and/or other phytosterols by multiple steroidogenic enzymes , epitomized by the members of the steroidogenic cytochrome P450 monooxygenases . High-level steroid hormone biosynthesis generally occurs in specialized steroidogenic tissues . Thus , an important condition for achieving tissue-specificity of steroid biosynthesis is providing a regulatory mechanism that ensures tissue-specific expression of the steroidogenic enzyme genes . In vertebrates , major sites of steroid hormone biosynthesis are the adrenal cortex , gonads and placenta , that express steroidogenic enzyme genes such as Cyp11a1 , P450c17a , 3β-HSD and 17β-HSD [1] . Key transcriptional regulators for these genes are the orphan nuclear receptors NR5A1 and NR5A2 , also known as Ad4BP/Steroidogenic Factor 1 ( SF-1 ) and Liver receptor homolog-1 ( LRH-1 ) , respectively [2–5] . Ad4BP/SF-1 and LRH-1 are predominantly expressed in the steroidogenic cells . A collective body of previous studies has established that Ad4BP/SF-1 controls steroid hormone biosynthesis through the transcriptional regulation of all steroidogenic genes [3 , 5] . Moreover , forced expression of this gene is sufficient to differentiate embryonic stem cells and human induced pluripotent stem cells into the steroidogenic cells [6 , 7] and to induce ectopic adrenal formation [8] , indicating that Ad4BP/SF-1 acts as a master regulator for steroid hormone biosynthesis in vertebrates . In insects , the principal steroid hormones are ecdysteroids , including ecdysone and its active derivative 20-hydroxyecdysone ( 20E ) , which plays pivotal roles in controlling a number of developmental and physiological events , especially in guiding transition from one developmental stage to the next via molting and metamorphosis [9–13] . During larval and pupal development , ecdysone is synthesized from dietary cholesterol in a specialized endocrine organ called the prothoracic gland ( PG ) . After release from the PG , ecdysone is converted to 20E in the peripheral tissues through the action of Shade , the terminal P450 monoxygenase in the biosynthetic pathway [14] . In the last 15 years , a number of genes encoding essential ecdysteroidogenic enzymes acting in the PG have been identified and characterized , including noppera-bo [15–17] , neverland ( nvd ) [18 , 19] , Cyp307a1/spook ( spo ) [20 , 21] , Cyp307a2/spookier ( spok ) [21] , non-molting glossy/shroud ( sro ) [22] , Cyp306a1/phantom ( phm ) [23 , 24] , Cyp302a1/disembodied ( dib ) [25 , 26] and Cyp315a1/shadow ( sad ) [26] . All of these enzymes ( except nvd and spok ) are collectively referred to as the Halloween genes [13 , 27] . Previous studies have identified multiple transcription factors essential for ecdysteroidogenic functions in the PG . For example , the Ecdysone receptor-Ultraspiracle complex and several other ecdysteroid-regulated transcription factors such as βFTZ-F1 , Broad , E75A and DHR4 are involved in both forward and feedback regulation of cyclic ecdysteroid production [28–35] . Ecdysteroid biosynthesis is also transcriptionally regulated by other factors including Without children [36] , Molting defective ( Mld ) [37] , the CncC-dKeap1 complex [38] , Ventral veins lacking ( Vvl ) [39 , 40] , Knirps [39] and FOXO [35] . Importantly , it has been reported that Broad , CncC , dKeap1 , Vvl and Knirps directly bind to enhancer regions of some ecdysteroidogenic enzyme genes [31 , 33 , 38 , 39] . However , it should be noted that , unlike vertebrate Ad4BP/SF-1 and LHR-1 , all of the identified steroidogenic transcription factors in insects are highly expressed not only in the PG , but also in many other non-ecdysteroidogenic tissues . Furthermore , some of these transcription factors have important functions other than ecdysteroid biosynthesis . For example , FOXO is well characterized as the primary transcriptional mediator of the insulin/insulin-like peptide signaling pathway in almost all cells [41] . Similarly , the CncC-dKeap1 complex is known to regulate xenobiotic responses [42] while Vvl and Knirps play key roles in cellular differentiation and morphogenesis of several tissues during embryogenesis including the PG ( i . e . [43 , 44] ) . More notably , βFTZ-F1 , the insect homolog of vertebrate Ad4BP/SF-1 , plays a crucial role in ecdysteroid-dependent transcriptional cascades in not only the PG but also many other tissues [4 , 9] . In contrast to the broad roles that all these steroidogenic factors play in other tissues during development , we describe here a much more specific role for the transcription factor coded by the gene ouija board ( ouib ) . Ouib is a C2H2-type zinc finger transcription factor , that is specifically expressed in the Drosophila PG and our genetic analysis clearly demonstrates that ouib is only essential for the expression of Spookier ( Spok ) , a potential rate-limiting enzyme in the ecdysone biosynthetic pathway . Most remarkable , however , is that spok appears to be the essential target of ouib since resupply of a Spok paralog in PG tissue rescues ouib mutants to viability . Since orthologs of ouib and spok are found only in Drosophiladae genomes , this study also suggests a presence of insect clade-specific transcriptional regulatory mechanisms of ecdysone biosynthesis .
We identified CG11762 , designated ouija board ( ouib ) , as a gene predominantly expressed in the PG primordia in the embryonic in situ gene expression pattern database of the Berkeley Drosophila Genome Project Experiment ID RT01107 [45] . We confirmed the PG restricted expression of ouib in embryos using RNA in situ hybridization ( Figs 1A , 1B and S1 ) . Additional RNA in situ hybridization and quantitative reverse-transcription ( qRT ) -PCR experiments revealed that ouib is also predominantly expressed in the ring gland including the PG cells during larval development ( Fig 1C , 1D and 1E ) . These results suggest that ouib may be involved in ecdysteroid biosynthesis . The predicted open reading frame of ouib encodes a protein that belongs to the family of the zinc-finger associated domain ( ZAD ) containing C2H2 zinc-finger proteins ( ZFPs ) [46 , 47] . The ZAD-ZFP family constitutes the largest subgroup of C2H2 ZFPs especially in insect species , and are characterized by an N-terminal ZAD consisting of ∼75 amino acid residues that are thought to serve as a protein-protein interaction domain [48] . In D . melanogaster , there are 98 independent loci encoding ZAD-ZFPs [47] . At least some of ZAD-ZFPs are thought to act as transcription factors , since several of them have been reported to bind DNA [49 , 50] . Notably , 5 paralogs of ouib are duplicated at the 85A9 cytological position of the third chromosome in D . melanogaster genome ( S2 Fig ) , and one of the paralogs designated M1BP codes for a general transcription factor [51] , raising the possibility that Ouib acts as a transcription factor in the PG . Orthologs of ouib are found in genomes of 11 other Drosphilidae species ( S1 Table ) [52] . FlyBase ( http://flybase . org/reports/FBgn0037618 . html ) also indicates the presence of potential orthologs of ouib in the mosquito species Aedes aegypti , Anopheles gambiae and Culex quinquefasciatus . However , a reciprocal BLAST search does not support the idea that the mosquito genomes have true ouib orthologs . In addition , a standard BLAST search did not detect any orthologous counterparts of ouib in any organisms other than Drosophilidae species . This is consistent with the previous report [47] , that no orthologs of ouib are found in genomes of the silkworm Bombyx mori and the beetle Tribolium castatenum . Taken together , these results suggest that ouib is a Drosophilidae-specific ecdysteroidogenic component . To assess the in vivo functional importance of ouib , we generated ouib loss-of-function alleles by a CRISPR/Cas9-dependent genome editing technology [53] . We succeeded in isolating two independent mutant alleles , ouib29 and ouib74 , each of which had a small deletion induced by different CRISPR single guide RNAs ( sgRNAs; Fig 2A and 2B ) . Both ouib29 and ouib74 alleles led to premature stop codons in the putative coding sequence of ouib , eliminating all 5 zinc-finger domains in the C-terminal region of Ouib ( Fig 2C ) . Embryos transheterozygous for ouib29/ouib74 completed embryogenesis , hatched normally , and showed no apparent morphological defects after hatching . However , ouib29/ouib74 transheterozygotes arrested development in the first instar larval stage and , even 108 hours after egg laying ( AEL ) or later , never molted into second instars and ( Fig 3A and 3B ) . Eventually all ouib29/ouib74 transheterozygous animals died by 144 hours AEL retaining the first instar larva-type morphology . In contrast , the majority of control ouib29/+ or ouib74/+ heterozygous animals became pupae ( Fig 3A and 3B ) by this time . To rule out the possibility that the observed phenotype was due to off-target mutations by CRISPR/Cas9 system , we combined ouib29 or ouib74 allele with a deficiency ( Df ) line that deletes a genomic region containing ouib locus . Similar to ouib29/ouib74 transheterozygotes , ouib29/Df or ouib74/Df animals died in the first instar stage , while +/Df animals were fully viable . This result provides evidence that ouib locus is responsible for the lethal phenotype . These results demonstrate that Ouib is essential for larval development . We next examined whether the larval arrest and lethality phenotype of ouib mutant animals was due to the loss of ecdysteroids . An ELISA assay revealed that the ecdysteroid titer in ouib29/ouib74 transheterozygotes was significantly reduced compared to control animals ( Fig 3C ) . Consistent with this observation , the expression of E75A , which is an early ecdysteroid-inducible gene , was greatly reduced in ouib29/ouib74 transheterozygotes ( Fig 3D ) . Moreover , when ouib29/ouib74 animals or ouib74/Df animals were fed yeast paste containing 20E after hatching , they molted to the second instar larval stage or later , as judged by the anterior spiracular morphologies ( Table 1 ) . These results suggest that loss of ouib mutant phenotype is due to ecdysteroid deficiency and that ouib regulates ecdysteroid production in the PG during normal development . As described above , we expect that Ouib acts as a transcription factor . Considering the spatial expression pattern and the loss-of-function phenotype of ouib , we wondered whether loss of ouib resulted in changes in the expression levels of any ecdysteroidogenic genes in the PG . To address this issue , we conducted qRT-PCR experiment to examine expression levels of 6 ecdysteroidogenic genes in the first instar larvae of control and ouib29/ouib74 transheterozygotes . Among the 6 genes , the expression of one gene Cyp307a2/spok was drastically reduced in ouib29/ouib74 transheterozygotes as compared to control animals ( Fig 4A ) . An immunohistological analysis using anti-Spok antibody also revealed a strong decrease of Spok protein level in ouib29/ouib74 larvae compared to control animals , but not that of the Sro protein , another ecdysone biosynthetic enzyme expressed in the PG . ( Fig 4B ) . We also found that expression of Cyp302a1/dib and Cyp315a1/sad , two other ecdysone biosynthetic P450 genes , were also lower than in ouib29/ouib74 animals compared to control animals , but their reduction was just on the threshold of significance ( Fig 4A ) . On the basis of the observation that the mutants cannot induce the expression of “spookier , ” we named CG11762 “ouija board” since this is an instrument for calling ghosts in western countries . We also examined whether there was a correlation between expression of ouib and spok during larval development . Overall the expression of both genes was relatively low during early stages and gradually became higher in late stages . We also found that the temporal expression profile of ouib closely correlates to that of spok in the late third instar stage ( Fig 4C ) . Curiously , the temporal expression profile did not always correlates to the dynamics of ecdysteroid titer during the third instar stage ( S3 Fig ) . For example , ouib expression did not increase prior to white prepupal stage , when the level of ecdysteroid titer was high . This result suggests that the ouib-spok coordinated transcriptional relationship does not fully account for the temporal dynamics of ecdysteroid biosynthesis during development . A previous study reported that Spok plays a crucial role in the “Black Box” , which consists of the conversion steps from 7-dehydrocholesterol ( 7dC ) to 5β-ketodiol ( 5βkd ) in the ecdysteroid biosynthetic pathway [21] . The participation of Spok in the “Black box” reactions was inferred by the observation that the larval arrest phenotype of spok RNAi animals was rescued by oral administration of 5βkd , but not 7dC or the most upstream precursor cholesterol [21] . Indeed , the same tendency as observed in spok RNAi animals was found in ouib loss-of-function animals . When ouib29/ouib74 transheterozygotes were fed yeast paste supplemented with cholesterol or 7dC , the larvae still arrested at the first instar larval stage ( Fig 4D and Table 1 ) . In contrast , we found that the first instar larval arrest phenotype of ouib29/ouib74 transheterozygotes was rescued when the animals were fed yeast paste supplemented with 5βkd ( Fig 4D and Table 1 ) . These results suggest that loss of ouib function specifically impairs the catalytic conversion that takes place during the “Black Box” reactions . These results also imply that the moderate reduction seen in dib and sad expression does not contribute in a major way to the ouib mutant phenotype . In addition to the feeding rescue experiment , we examined whether the ouib mutant phenotype was rescued by forced expression of spok using GAL4-UAS binary gene expression system . We first established UAS-spok transgenic strains to drive spok expression in the PG cells under control of phm-GAL4#22 driver . However , for an unknown reason , none of our UAS-spok transgenes was expressed in the PG of first and second instar larvae with the phm-GAL4#22 driver and thus these constructs were not suitable for our experimental purpose . Therefore , we decided to examine whether the ouib mutant phenotype could be rescued by forced expression of Cyp307a1/spo , a paralog of spok that appears to provide the same enzymatic activity but only in embryos and in the follicular cells of the ovary [20 , 21] . We confirmed that spo was functionally equivalent to spok in vivo , as spo overexpression rescued the first instar larval arrest phenotype of spok RNAi animals ( S2 Table ) . Indeed , spo overexpression in the PG rescued the larval arrest phenotype of ouib29/ouib74 transheterozygotes , and some of the animals grew up to the adult stage ( Table 2 ) . These results strongly suggest that the developmental arrest phenotype of ouib mutant is due solely to loss of spok expression in the PG . Our data therefore support the idea that Ouib is a special transcription factor primarily required for inducing expression of one biosynthetic gene spok , and no other essential gene during development . To address whether Ouib protein acts directly on the spok enhancer region to induce spok expression , we initially searched for an Ouib-response element in the enhancer/promoter region of spok . We first identified a ~1 . 4 kb genomic region upstream of the spok coding sequence that was sufficient to mimic the expression of spok in the PG when fused to a GFP reporter ( Figs 5A and S4 ) . The GFP expression driven by the ~1 . 4 kb spok enhancer region was almost completely abolished in ouib29/ouib74 transheterozygotes ( Fig 5A ) , suggesting that the ~1 . 4 kb element contains a Ouib-response element . In order to identify the cis-regulatory element ( s ) responsible for the Ouib-mediated control of spok expression , we conducted a promoter/enhancer characterization analysis in a heterologous cell culture system . We generated DNA constructs carrying the upstream region of spok fused with a luciferase ( luc ) gene cassette and then transfected Drosophila Schneider 2 ( S2 ) cells using these DNA constructs with or without a plasmid for overexpressing FLAG-ouib . We identified a 300 bp genomic region corresponding to the region from -331 bp to -32 bp upstream of the ATG start codon of spok that drives expression of the luc reporter in S2 cells in an Ouib-dependent manner ( S5 Fig ) . The 300 bp region was also sufficient to drive expression of a GFP reporter in the PG cells ( S4 Fig ) . To narrow down the element ( s ) responsible for the Ouib-dependent expression of spok , we tested several constructs carrying the upstream region of spok with a range of deletions within the 300 bp region ( S5 Fig ) . We first generated the deletion constructs in 50 bp increments from 5´ terminus of the 300 bp region and found that the region from -181 to -131 bp was crucial for the Ouib-dependent luc reporter activity ( S5 Fig ) . We then generated the deletion constructs in 10 bp increments from 5´ terminus of the -181 to -32 region . The construct carrying the -151 to -32 region did not show any induction in luc reporter activity even in the presence of Ouib ( Fig 5B ) . The construct carrying the longer 10 bp 5´ extension ( -161 to -32 ) still retained statistical significant Ouib-dependent luc reporter activity . However , the fold induction of luc reporter activity with the -161 to -32 region was slightly reduced as compared to the -171 to -32 region or longer ( Fig 5B ) . From these results , we hypothesized that the Ouib-response element lay between -166 to -152 bps ( Fig 5C ) . To clarify the importance of this 15 bp region for Ouib-dependent control of gene expression , we introduced transversion mutations of the entire 15 bp sequence . This mutated construct exhibited no luc reporter induction in the presence of Ouib upon transfection into S2 cells ( Fig 5D ) . We also conducted subsequent reporter assays using constructs carrying various mutations in the 15 bp sequence . None of the constructs carrying any of several 3 bp substitutions within the 15bp sequence eliminated the responsive to Ouib ( S6 Fig ) . Therefore , we conclude that Ouib binding tolerates degeneracy throughout the 15 bp sequence ( 5´-AGCTTTATTATTTAG-3´ ) . We also examined the evolutionary conservation of the Ouib-response elements in putative spok enhancer regions in 12 Drosophilide species whose genome sequences have been determined [52] . EMBOSS Matcher , an algorithm to identify local similarities between two sequences [54] , found sequence motifs similar to the D . melanogster Ouib-response element in almost all of the Drosophilidae species ( S7 Fig ) . In particular , the D . yakuba putative spok enhancer contains exactly the same 15 bp sequence motif . In addition , in the species belonging to the subgenus Sophophora , which includes D . melanogaster , the Ouib-response element-like motifs are found in proximity ( within 500 bp ) to the spok coding region ( S7 Fig ) . These data suggest that Ouib-like response elements are also evolutionarily conserved to some degree . We sought to further establish if Ouib binds directly to the Ouib-response element by performing a DNA/protein binding assay . We first examined the physical interaction between the Ouib-response element sequence and Ouib protein by an ABCD assay , which uses biotin conjugated , double-stranded oligonucleotides containing the Ouib-response element sequences . Nuclear extracts obtained from S2 cells expressing FLAG-ouib were mixed with the biotin-labeled oligonucleotide , and then the protein-oligonucleotide complexes were pulled down using streptavidin beads . We found that FLAG-Ouib protein bound strongly to the wild type Ouib-response element probe , but not to the mutated probe ( Fig 6A ) . In the control experiments , a biotinylated probe corresponding to M1BP ( another ZAD-ZFP homolog of Ouib ) binding sequence in the enhancer of smoothened locus [51] did not efficiently precipitate FLAG-Ouib . Conversely , FLAG-M1BP protein did not bind to the Ouib-response element , while it bound to M1BP binding element ( Fig 6A ) . To exclude the possibility that FLAG-Ouib protein isolated from cultured cells , indirectly associated with the probe through a complex containing some other endogenous transcription factor unrelated to Ouib , we prepared an E . coli produced recombinant protein containing the C-terminal 5 zinc finger domains ( Ouib-Zf ) , and performed electrophoretic mobility shift assays ( EMSAs ) between the recombinant protein and the 15 bp Ouib-response element . We utilized 45 bp radiolabeled DNA probes , whose sequences corresponded to the spok enhancer region containing the 15 bp Ouib-response element . We found that the wild type oligonucleotide probes formed DNA/protein complexes with GST-Ouib-Zf , but not with GST alone ( Fig 6B and 6C ) . In contrast , such DNA/protein complexes were not detected when radiolabeled mutated Ouib-response element sequences or sequences corresponding to the M1BP site [51] ( Fig 6D ) were used as probes , thereby confirming the specificity of the binding . Moreover , the complexes with the wild type 45 bp probes were outcompeted by unlabeled 45 bp DNA probes with the wild type Ouib-response element sequences , but not by the unlabeled mutated DNA probes or by the unlabeled M1BP probes ( Fig 6B and 6C ) . Taken together , these findings strongly support the idea that Ouib specifically regulates spok transcription by direct binding to the Ouib-response element in the spok enhancer . Previous studies found that the increase of spok expression in the late third instar larvae is positively controlled by prothoracicotropic hormone ( PTTH ) [55 , 56] . We therefore examined whether ouib expression changed in response to down regulation of PTTH signaling . However , when the levels of the PTTH receptor gene torso were knocked down in the PG by RNAi , we observed no change in ouib expression , suggesting that PTTH regulation of spok is not mediated through ouib ( S8 Fig ) .
In this study , we have demonstrated that the ZAD-ZFP Ouib is required for ecdysteroid biosynthesis in the PG during D . melanogaster development . The following points summarize our finding . First , ouib is predominantly expressed in the PG during embryonic and larval stages . Second , ouib null mutants exhibit early ( first instar ) larval developmental arrest due to a low ecdysteroid titer . Third , the larval arrest phenotype is caused by a failure of spok expression in the PG , and is rescued by sole overexpression of a spok paralog . Finally , a specific Ouib-response element that binds Ouib was identified in the enhancer region of spok . Our study reports on the discovery of the first invertebrate tissue-specific , steroidogenic transcription factor . ouib mutants exhibit a drastic reduction of spok expression . However , we point out that ouib mutants also show a mild statistically-significant reduction of dib and sad ( Fig 4A ) . In fact , while no DNA sequences exactly matching the spok Ouib-response element ( 5´-AGCTTTATTATTTAG-3´ ) are found elsewhere in D . melanogaster genome , a number of degenerate sequences do exist in the genome , including the regions upstream of ecdysteroidogenic gene coding regions ( S9 Fig ) . Considering the fact that the luciferase constructs carrying any of several 3 bp substitutions within the 15bp sequence are still responsive to Ouib ( S6 Fig ) , we cannot completely rule out the possibility that Ouib is also involved in direct transcriptional regulation of genes other than spok , particularly dib and sad . Nevertheless , our results indicate that the impairment of expression of dib and sad seems not to contribute to the ouib phenotype in a major way . First , the arrest during the first instar larval stage of ouib mutants is rescued by oral administration of 5βkd . Since Dib and Sad play roles in the terminal hydroxylation steps downstream of the conversion of 5βkd to ecdysone [23 , 24 , 26] , this finding suggests that the enzymatic levels of Dib and Sad are still sufficient to make functional levels of ecdysone . Second and more importantly , the first instar larval arrest phenotype of ouib mutants is rescued by a sole overexpression of spo , which is functionally equivalent to spok . Therefore , in addition to the PG specificity , we would argue that the key additional feature of Ouib is its specific role in spok expression . To further clarify the extent to which Ouib regulates other genes , and the functional importance of these genes , will require additional studies including transcriptome analysis/ChIP-seq analysis together with eventual mutational analysis of any identified targets . Curiously , the presence of ouib only in the Drosophilidae genomes is concordant with the Drosophilidae-specific duplication of Cyp307a P450 subfamily . While members of the Cyp307 P450 subfamily , which includes spok , are found in all arthropod species examined so far [20 , 21 , 57 , 58] , Drosophilidae Cyp307 genes have been duplicated within the Drosophila radiation [21 , 59] . In the case of D . melanogaster , the duplicated Cyp307 genes are Cyp307a1/spo and Cyp307a2/spok , which are sub-functionally divergent in terms of gene expression pattern; spo is expressed in early embryogenesis and oogenesis , while spok is expressed in the PG cells in late embryogenesis as well as the larval and pupal stages [20 , 21] . Our data demonstrate that the spatiotemporal expression pattern of ouib closely matches that of spok but not spo . Notably , neither ouib nor spok transcripts are detected in embryonic stages 5–9 when the embryonic ecdysteroid titer is maximal [21 , 45] , indicating that these genes do not contribute to producing embryonic ecdysteroids . Therefore , an acquisition of ouib might be a critical event for the sub-functionalization of two Cyp307 genes by changing the regulation of their expression during the Drosophilidae evolution . In terms of evolution , it is worth mentioning that there is a case where evolution changed the activity of a single ecdysteroidogenic enzyme ( Nvd ) dramatically limiting the food source of Drosophila pachea to a single species of cactus [60] . Further assessment of the biological and evolutionary roles of ouib and spok will require determining which transcription factors are involved in the transcriptional regulation of D . melanogaster spo and Cyp307a genes in other insects . Since there are many divergent ZAD-ZFP genes in each insect genome and they are expanded in insect lineage-specific manner [47] , it is possible that a different ZAD-ZFP gene whose primary structure is not orthologous to ouib could be a transcription factor for other Cyp307a genes . Regarding the evolutionarily aspect of ouib , it is important to recognize that spok expression is regulated by another ZAD-ZFP called Molting defective ( Mld ) [21 , 37 , 39] . Interestingly , just like ouib , mld genes are also found only in genomes of Drosophilidae but not other insects [21 , 37] . In contrast to ouib , Mld does not appear to be specific for the regulation of spok expression . First , Mld , unlike ouib , is expressed in several other tissues during development besides the PG [37] . Second , Mld is essential for regulating expression Nvd as well as spok and perhaps other genes [39] . Third and most important , the mld loss–of-function phenotype is not rescued by overexpressing either spo or spok [21] . Therefore , Ouib and Mld overlap in their regulation of spok expression , but also have distinct functions during development . While it is still unclear whether Mld is a transcription factor , it would be intriguing to examine a functional relationship between Ouib and Mld for induction of spok expression in the PG . According to our qRT-PCR data , it is less likely that Mld controls ouib expression in the PG ( S10 Fig ) . Another question to be answered is how ouib expression is regulated during larval development . As shown above , it does not seem to be by PTTH . However , recent work as shown that spok and other ecdysteroidogenic enzyme genes are also influenced by humoral factors such as TGFβ/Activin [61] and monoaminergic tropic factors [62 , 63] . It will be interesting to determine whether these factors affect spok expression in the PG through modulation of ouib levels . An additional significant aspect of this work is to provide the first evidence for the existence of a catalytic step-specific transcriptional regulation of steroid hormone biosynthesis in organisms . Whereas the substrate of Spok and its product have not yet been identified , Spok appears to play a crucial role in the “Black Box” step of ecdysteroid biosynthetic pathway , and it is a strong candidate for acting as a rate-limiting enzyme in the pathway [10 , 21 , 64] . Interestingly , a recent study has reported that pre-mRNA splicing of spok , but not any other ecdysteroidogenic genes expressed in the PG , seems to specifically depend on a protein encoded by ecdysoneless ( ecd ) , whose mutant phenotype includes ecdysteroid deficiency [65] . Thus , a rate-limiting step of ecdysteroid biosynthesis catalyzed by Spok could be under tight control by both specific transcriptional and post-transcriptional mechanisms . Currently , it is unknown whether such catalytic-specific transcriptional and/or posttranscriptional mechanisms also exist in other organisms including vertebrates . Similar to ecdysteroids , vertebrate steroid hormones are synthesized via several intermediates by multiple steroidogenic enzymes . Among them , the rate-limiting step in vertebrate steroid hormone productions is the delivery of substrate cholesterol from the outer mitochondrial membrane to the inner one and the subsequent conversion of cholesterol to pregnenolone by CYP11A1 . It is attractive to hypothesize that the rate-limiting step in vertebrate steroid hormone biosynthesis is also specifically regulated by unidentified transcriptional and/or splicing regulator ( s ) . Whereas no apparent orthologs of ouib are found in vertebrates , their genomes possess a ZAD-ZFP gene called ZFP276 , which is a tumor suppressor gene [66] . Interestingly a ecd ortholog is also found in humans and may also contribute to the malignancy of certain tumor types [65] . It would be worth examining roles of these genes in steroid hormone biosynthesis in vertebrates .
Drosophila melanogaster flies were reared on standard agar-cornmeal medium at 25°C under a 12:12 h light/dark cycle . w1118 , yw and Oregon R were used as the wild type strain . phm–GAL4#22 [55] and w; UAS-dicer2; phm-GAL4#22/TM6 Ubi-GFP was used as the strain to drive forced gene expression in the PG . UAS-spo [20] and UAS-spok-IR [21] transgenic flies were obtained from Hiroshi Kataoka ( The University of Tokyo ) and Hajime Ono ( Kyoto University ) , respectively . y1 v1 nos-phiC31; attP40 , v1 and y2 cho2 v1; attP40{nos-Cas9}/CyO [53] were obtained from National Institute of Genetics , Japan . The w; snaSco/CyO; P{w+mC = tubP-GAL80ts}7 ( stock number #130453 ) and w1118; Df ( 3R ) ED5330/TM6C Sb1 , a deficiency strain that deletes a genomic region including the ouib locus ( stock umber #150241 ) [67] , were obtained from Drosophila Genetic Resource Center . UAS-torso-IR ( stock number #101154 ) and UAS-mld-IR ( stock number #17329 ) were obtained from the Vienna Drosophila RNAi center . Digoxygenin ( DIG ) -labeled antisense RNA probes were synthesized using DIG RNA labeling mix ( Roche ) and T3 and T7 RNA polymerase ( Fermentas ) . To generate the ouib probe , the ouib ORF was amplified by PCR with cDNA derived from whole bodies of Oregon R larvae and the primers described in S3 Table . PCR product was inserted into SmaI-digested pBluescript II SK ( - ) , and then used as the templates for synthesizing RNA probes . Fixation , hybridization and detection were performed as [23 , 68] . RNA was isolated using the RNAiso Plus reagent ( TaKaRa ) . Genomic DNA digestion and cDNA synthesis were performed using the ReverTra Ace qPCR RT Kit ( TOYOBO ) . qRT-PCR was performed using the THUNDERBIRD SYBR qPCR Mix ( TOYOBO ) or Universal SYBR Select Master Mix ( Applied Biosystems ) with a Thermal Cycler Dice TP800 or TP870 system ( TaKaRa ) . Serial dilutions of a plasmid containing the ORF of each gene were used as a standard . The expression levels of the target genes were normalized to an endogenous control ribosomal protein 49 ( rp49 ) in the same sample . The primers for quantifying D . melanogaster ouib and E75A are described in S3 Table . Primers amplifying nvd , sro , spok , phm , dib , sad and rp49 were previously described [22 , 55] . Tissue dissections were performed in PBS followed by fixation in 4% PFA for 20 minutes at room temperature . For this study , the following primary antibodies were: mouse anti-FLAG M5 ( 1:1 , 000 ) ( Sigma ) ; rabbit anti-Phm ( 1:200 ) [30] , guinea pig anti-Spok ( 1:200 ) [61]; guinea pig anti-Sro ( 1:1 , 000 ) [62] . Tissues were incubated over night with primary antibodies at 4°C . Fluorescent conjugated secondary antibodies used in this study , goat anti-mouse Alexa Fluor 488 , goat anti-guinea pig Alexa Fluor 488 , goat anti-rabbit Alexa Fluor 555 and goat anti-guinea pig Alexa Fluor 555 , were purchased from Life Technologies . Secondary antibodies were diluted 1:500 and incubated for 1 hour at room temperature . Confocal images were captured using Carl Zeiss LSM 700 laser scanning microscope . The GAL4-UAS system [69] was used to overexpress genes in D . melanogaster . To generate pUAST vector to overexpress ouib , specific primers including a sequence coding FLAG tag at N terminal were used for PCR to add EcoRI and XbaI sites to the 5´ and 3´ ends , respectively ( S3 Table ) . Template cDNAs were reverse transcribed using total RNA of the ring gland from D . melanogaster using ReverTra Ace qPCR RT Kit ( TOYOBO ) . PCR was performed using KOD Plus Neo ( TOYOBO ) . The amplified CDS region of ouib was digested with EcoRI and XbaI , and then ligated into a pWALIUM10-moe vector [70] . Transformants were established by BestGene , Inc . Generation of the ouib allele was carried out by CRISPR/Cas9 system using the pBFv-U6 . 2 vector [53] provided by the National Institute of Genetics , Japan . We selected 2 independent target sites ( target#1 and target#2 as shown in Fig 2 ) . To minimize off-target effects of CRISPR/Cas9 system , we confirmed by BLAST search that no 15 nucleotide stretches within the selected target sequence ( 23 nucleotides including PAM motif ) matched any other sequence on the 3rd chromosome . Sense and antisense oligonucleotides corresponding to sgRNA target sequences ( S3 Table ) were annealed and inserted into BbsI-digested pBFv-U6 . 2 vector . The ouib sgRNA vectors were injected into the embryos of the y1 v1 nos-phiC31; attP40 strain . The nos-Cas9-based gene targeting was carried out as previously described [53] . Males carrying nos-Cas9 and a sgRNA transgene were crossed to wild-type flies by mass mating . From their progeny , 10 and 50 single males for the target#1 and target#2 sites , respectively , were isolated . Each male was crossed with w; TM3 Sb/TM6 Tb females and then the independent isogenized strains were established . Among them , we surveyed the strains showing homozygous lethality and eventually 1 target#1 and 29 target#2 lethal strains were selected . To confirm indel mutations at ouib locus in each strain , we performed the T7EI assay as previously described [53] . In this assay , genome DNA from the heterozygous adults of each strain was extracted as previously described [53] . To amplify the DNA fragment including Cas9 target sites , PCR was conducted with KOD FX Neo ( TOYOBO ) , the extracted genome DNA , and the primers listed in S3 Table [53] . The PCR products were treated with T7 endonuclease ( NEB ) . The reacted samples were analyzed by agarose gel electrophoresis . Out of 30 total candidate strains , 1 target#1 and 8 target#2 strains were selected as candidate flies possessing indel mutations in ouib region . The PCR products from the 9 strains were subcloned into a SmaI-digested pBluescript II ( Promega ) and then sequenced with T3 and T7 primers . We detected small deletions in 8 out of the 9 strains . The minimal and maximal deletion sizes were 1 bp and 13 bp , respectively . We chose 1 strain for each target sites for further analyses and renamed them ouib29 and ouib74 , both of which caused frameshift mutations for ouib locus ( Fig 2 ) . ouib29/TM3 Act-GFP flies , ouib74/TM3 Act-GFP flies and w1118 flies were crossed each other . Eggs were laid on grape plates with yeast pastes at 25°C for 8 hours . 36 hours AEL , 100 hatched GFP negative ( ouib29/+ , ouib74/+ and ouib29/ouib74 ) first instar larvae were transferred into vials with standard cornmeal food ( 25 animals per vial ) . Every 24 hours , developmental stages were scored by tracheal morphology as previously described [22] . For the rescue experiments , 20 mg of dry yeast was mixed with 38 μl H2O and 2 μl ethanol or supplemented with 2 μl of the following sterols dissolved in ethanol: cholesterol ( Wako; 150 mg/ml ) , 7-dehydrocholesterol ( Sigma; 150 mg/ml ) , 5β-ketodiol ( kindly gifted from Yoshinori Fujimoto , Tokyo Institute of Technology; 150 mg/ml ) and 20-hydroxyecdysone ( Sigma; 50 mg/ml ) . We crossed ouib29/TM3 Ser1 GMR2 Act-GFP flies with ouib74/TM3 Ser1 GMR2 Act-GFP flies . Eggs were laid on grape plates with yeast pastes at 25°C for 12 hours . At 36 hours AEL , 50 hatched GFP negative ( ouib29/ouib74 ) first instar larvae were transferred to the yeast paste on grape plates and kept at 25°C . Every 24 hours , developmental stages were scored by tracheal morphology as previously described [22] . For the rescue experiments of ouib mutant by ouib overexpression , ouib29 phm-GAL4#22/TM3 Act-GFP was established by chromosomal recombination . The flies of UAS-FLAG-ouib-1M; ouib74/TM6 Ubi-GFP were crossed with the flies of ouib29 phm-GAL4#22/TM3 Act-GFP , the flies of ouib74/TM3 Act-GFP were crossed with the flies of ouib29 phm-GAL4#22/TM3 Act-GFP , and the flies of UAS-FLAG-ouib-1M; ouib74/TM6 Ubi-GFP were crossed with the flies of ouib29/TM3 Act-GFP . Eggs were laid on grape plates with yeast pastes at 25°C for 12 hours . At 36 hours AEL , 50 hatched GFP negative ( UAS-FLAG-ouib-1M/+; ouib29 phm-GAL4#22/ouib74 , ouib29 phm-GAL4#22/ouib74 and UAS-FLAG-ouib-1M/+; ouib29/ouib74 ) first instar larvae were transferred to the standard agar-cornmeal medium . Developmental stages were scored 108 hours AEL by tracheal morphology as previously described [22] . For the rescue experiments of spok RNAi by spo overexpression , UAS-spok-IR UAS-spo was established by chromosomal recombination on third chromosome . The flies of UAS-spok-IR UAS-spo strain was crossed with w; UAS-dicer2; phm-GAL4#22/TM6 Ubi-GFP flies . Eggs were laid on standard agar-cornmeal medium at 25°C for 24 hours . After 7 days , developmental stages of the animals on the wall were scored by presence of TM6 balancer . For the rescue experiments of ouib mutant by spo overexpression , Roi/CyO; ouib29 phm-GAL4#22/TM6 , Roi/CyO; ouib29 UAS-spo/TM6 and Roi/CyO; ouib74 UAS-spo/TM6 were established by chromosomal recombination on third chromosome . The flies of Roi/CyO; ouib29 phm-GAL4#22/TM6 were crossed with Roi/CyO; ouib74 UAS-spo/TM6 , the flies of Roi/CyO; ouib29 phm-GAL4#22/TM6 were crossed with Roi/CyO; ouib74/TM6 and Roi/CyO; ouib74/TM6 were crossed with Roi/CyO; ouib29 UAS-spo/TM6 . Eggs were laid on standard agar-cornmeal medium at 25°C for 24 hours . After 7 days , developmental stages of the animals on the wall were scored by presence of TM6 balancer . ouib29/TM3 Ser1 GMR2 Act-GFP flies and w1118 flies were crossed with ouib74/TM3 Ser1 GMR2 Act-GFP flies . Eggs were laid on grape plates with yeast pastes at 25°C and the hatched larvae were cleared . After 8 hours , GFP negative ( ouib74/+ and ouib29/ouib74 ) first instar larvae were transferred into vials with standard cornmeal food . At 12 hours AH , whole larvae were rinsed in water and homogenized in 50 μl methanol and supernatant was collected following centrifugation at 14 , 000 rpm at 4°C . The remaining tissue was re-extracted in 50 μl methanol over night at 4°C . The supernatants were evaporated using a EYELA CVE-2000 ( Tokyo Rikakikai ) and redissolved in 50 μl EIA buffer [0 . 1 M PBS/0 . 1% BSA , 0 . 4 M NaCl , 1 mM EDTA and 0 . 01% NaN3] . ELISA was performed according to manufacturer’s instructions using 20-Hydroxyecdysone EIA Antiserum , 20-Hydroxyecdysone AChE Tracer and Ellman’s Reagent ( Cayman Chemical ) that detects 20-hydroxyecdysone with the same affinity . Standard curves were generated using 20E ( Sigma ) . Absorbance was measured at 415 nm on a plate reader , Multiskan GO ( Thermo Scientific ) using the SkanIt Software 3 . 2 ( Thermo Scientific ) . To generate the spok>GFP reporter construct , a ~1 . 4 kb fragment immediately upstream of the spok transcription unit was amplified from yw genomic DNA using the primers 1 . 45spok-p_F and 1 . 45spok-p_R ( S3 Table ) . This fragment was first subcloned into the pCR2 . 1-TOPO vector ( Life Technologies ) and then removed as an EcoRI fragment and cloned into the Drosophila transformation vector pH-Stinger [71] . To refine the location of the PG enhancer , seven 250–300 bp overlapping fragments that covered the entire 1 . 4 kb fragment were derived through PCR and each cloned into hH- Stinger . The only fragment that gave expression in the PG of transgenic animals was the ~300 bp fragment immediately upstream of the transcriptional start site . This fragment was generated using the primers 300spok-p_F and 300spok-p_R ( S3 Table ) . Transgenic lines were generated through standard means using a w1118 host background . The GFP reporter strains of spok>GFP; ouib29/TM6 Ubi-GFP and spok>GFP; ouib74/TM6 Ubi-GFP were established and crossed each other . Eggs were laid on grape plates with yeast pastes at 25°C for 4 hours . The first instar larvae were dissected 36 hours AEL and immunostained . The upstream regions of spok were amplified from Oregon R genomic DNA by specific primers to add SacI and BglII sites to the 5´ and 3´ ends , respectively . PCR was performed using KOD Plus Neo ( TOYOBO ) . The amplified upstream regions of spok were digested with SacI and BglII , and then ligated into a pGL3-Basic vector luciferase reporter plasmid ( Promega ) . Reporter plasmids carrying mutated regions were constructed from the pGL3-Basic plasmid containing WT upstream 300 bp region by inverse PCR . The primers for PCR are listed in S3 Table . S2 cells were seeded in 1 ml Schneider’s Drosophila Medium ( GIBCO ) in a 24-well plate ( greiner bio-one ) 1 day before transfection . Transfection of S2 cells was performed using the Effectene Transfection Reagent ( Qiagen ) . GFP-pUAST [23] and FLAG-ouib-pWALIUM10-moe plasmids were transfected , respectively , along with the Actin5C-GAL4 construct ( a gift from Yasushi Hiromi , National Institute of Genetics ) and the luciferase reporter plasmids . The Copia Renilla Control plasmid ( addgene; #38093 ) [72] was used as the reference . The cells were incubated for 2 days after transfection . Then they were processed by using the Dual-Luciferase Reporter Assay System ( Promega ) in accordance with the manufacturer’s instructions and were analyzed with Flash’n glow LB 955 ( Berthold Technologies ) . S2 cells overexpressing FLAG-ouib or FLAG-M1BP were collected and washed with TBS . Cells were then centrifuged at 4000 g at 4°C for 5 min . The pellet was suspended and vortexed with 400 μl Buffer A [10 mM Hepes pH 7 . 9 , 10 mM KCl , 1 mM DTT and 1 unit Complete Mini ( Roche ) ] and 25 μl 10% NP-40 . Then sample was centrifuged at 1500 g at 4°C for 5 min . The pellet was suspended and vortexed with 50 μl Buffer C [20 mM Hepes pH7 . 9 , 400 mM NaCl , 2 mM MgSO4 , 1mM DTT and Complete Mini ( Roche ) ] , then shaked at 4°C for 30 min . After shaking , sample was centrifuged at 14 , 000 rpm at 4°C for 5 min and supernatant was collected . Samples were boiled with SDS sample buffer [150 mM Tris-HCl pH 6 . 8 , 0 . 6% SDS , 15% glycerol , 0 . 009 mg/μl Bromophenol blue , 5% 2-mercaptoethanol and 1 unit Complete Mini ( Roche ) ] for 5 min , and loaded on 12% polyacrylamide gel followed by transfer onto PVDF membrane ( GE Healthcare ) . Anti-FLAG M5 monoclonal antibody ( 1:1 , 000; Sigma ) was used for primary antibody and ECL Peroxidase labeled anti-mouse antibody ( 1:10 , 000; GE Healthcare ) was used for secondary antibody . The band was detected by ECL Ultra Lumigen TMA-6 ( GE Healthcare ) and Ez-Capture MG ( ATTO ) . Preparation of S2 cell nuclear extracts is described in the Supplemental Materials . ABCD assay was conducted essentially as previously described [73] . Biotin-labeled DNA probes were purchased from Life Technologies . The probes were incubated with Dynabeads M-280 Streptavidin ( Life Technologies ) at room temperature for 15 min . DNA-beads complexes were mixed with nuclear extracts and ABCD Binding Buffer [50 mM Hepes pH 7 . 9 , 150 mM NaCl , 0 . 5% Triton X-100 , 20 ng/μl poly ( dI/dC ) ] , and incubated at 4°C for 1 hour . After incubation , the beads were washed with ABCD Binding Buffer . The biotin-labeled oligonucleotides are listed in S3 Table . GST proteins fused with or without 150–313 amino acid residues of Ouib ( GST-Ouib-Zf ) containing 5 zinc finger domains were expressed using pGEX-4T-3 vector system ( GE Healthcare ) in Escherichia coli BL-21 strain . E . coli cells were harvested and crashed with sonication . GST alone and GST-Ouib-Zf were purified from the supernatant with AKTA start equipped with GSTrap affinity column ( GE Healthcare ) . Electrophoretic mobility shift assay was conducted as previously described [74 , 75] . 45 bp double-stranded oligonucleotide probes containing wild type M1BP binding site , wild type and mutated ( transversion ) Ouib response element were prepared by annealing single-strand oligonucleotides listed in S3 Table . The wild type M1BP binding site was derived from the smoothened promoter [51] . Double-stranded DNA fragment was end-labeled by using T4 polynucleotide kinase ( TOYOBO ) and [γ-32P]ATP . GST or GST-Ouib fusion proteins ( 400 ng ) were incubated for 30 min at 4°C in the reaction mixture [12 mM Hepes , pH 7 . 9 , 1 mM dithiothreitol , 1 mM EDTA , 60 mM KCl , 4 mM MgCl2 , 2 mM ZnSO4 , 50 ng/ul poly ( dI-dC ) , 1 mg/ml BSA and 12% Glycerol] in the presence or absence of 100–200-fold molar excess of specific double-stranded competitor DNA . A radiolabeled DNA probe ( 0 . 3 ng , 40 , 000 cpm ) was added , and the incubation was continued for 20 min at 4°C . The incubation mixture was directly loaded on a 5% non-denaturing polyacrylamide gel in 1 × TBE buffer [89 mM Tris-HCl , pH 8 . 0 , 89 mM boric acid , and 2 mM EDTA] , and electrophoresed at 4°C with buffer circulation . The gels were dried and analyzed with a bio-imaging analyzer Typhoon 8600 ( Amersham Pharmacia Biotech Inc ) . The competitor oligonucleotides used are listed in the S3 Table . | Steroid hormones are crucial for development and reproduction in multicellular organisms . The spatially-restricted expression of almost all steroid biosynthesis genes is key to the specialization of steroid producing cells . In the last decade , insects have become the focus for research on the biosynthesis of the principal steroid hormones , ecdysteroids . However , the transcriptional regulatory mechanisms controlling the ecdysteroid biosynthesis genes are largely unknown . Here we show that a novel zinc finger transcription factor Ouija board ( Ouib ) is essential for activating the expression of one ecdysteroid biosynthesis gene , spookier , in the ecdysteroid producing cells . Ouib is the first invertebrate transcription factor that is predominantly expressed in the steroidogenic organs and essential for development via inducing expression of the steroidogenic gene . In addition , this is the first report showing the catalytic step-specific control of steroid hormone biosynthesis through transcriptional regulation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | The Drosophila Zinc Finger Transcription Factor Ouija Board Controls Ecdysteroid Biosynthesis through Specific Regulation of spookier |
Dengue is a major public health problem in tropical and subtropical countries . Exploring the relationships between virological features of infection with patient immune status and outcome may help to identify predictors of disease severity and enable rational therapeutic strategies . Clinical features , antibody responses and virological markers were characterized in Vietnamese adults participating in a randomised controlled treatment trial of chloroquine . Of the 248 patients with laboratory-confirmed dengue and defined serological and clinical classifications 29 ( 11 . 7% ) had primary DF , 150 ( 60 . 5% ) had secondary DF , 4 ( 1 . 6% ) had primary DHF and 65 ( 26 . 2% ) had secondary DHF . DENV-1 was the commonest serotype ( 57 . 3% ) , then DENV-2 ( 20 . 6% ) , DENV-3 ( 15 . 7% ) and DENV-4 ( 2 . 8% ) . DHF was associated with secondary infection ( Odds ratio = 3 . 13 , 95% CI 1 . 04–12 . 75 ) . DENV-1 infections resulted in significantly higher viremia levels than DENV-2 infections . Early viremia levels were higher in DENV-1 patients with DHF than with DF , even if the peak viremia level was often not observed because it occurred prior to enrolment . Peak viremias were significantly less often observed during secondary infections than primary for all disease severity grades ( P = 0 . 001 ) . The clearance of DENV viremia and NS1 antigenemia occurs earlier and faster in patients with secondary dengue ( P<0 . 0001 ) . The maximum daily rate of viremia clearance was significantly higher in patients with secondary infections than primary ( P<0 . 00001 ) . Collectively , our findings suggest that the early magnitude of viremia is positively associated with disease severity . The clearance of DENV is associated with immune status , and there are serotype dependent differences in infection kinetics . These findings are relevant for the rational design of randomized controlled trials of therapeutic interventions , especially antivirals .
Dengue viruses ( DENVs ) are members of the Flavivirus genus and are the most important arboviral pathogens of humans . The four DENVs are antigenically-related and have single-stranded , positive-sense RNA genomes that share 60–70% sequence identity between each others [1] . There are no licensed vaccines to prevent dengue and vector control remains the cornerstone of public health interventions . The clinical outcome from DENV infection ranges from the asymptomatic to an acute , often debilitating illness called dengue fever ( DF ) , to the severe and potentially life-threatening dengue hemorrhagic fever ( DHF ) and dengue shock syndrome ( DSS ) . The cardinal feature of DHF/DSS is a capillary permeability syndrome characterised by plasma leaking from the vasculature into interstitial spaces . Thrombocytopenia , a coagulopathy and a hemorrhagic diathesis are also common findings . DSS manifests when capillary permeability is severe enough to result in an inadequate intravascular volume that then leads to poor tissue perfusion . DSS is managed , and possibly prevented , by careful restoration and maintenance of the intravascular volume by use of parenteral fluids . Viral strain and host immune status have been suggested as major risk factors for DHF/DSS . In particular , two sequential infections , with the second infection caused by a DENV serotype different from the first , is a risk factor for severe disease in children and adults [2]–[5] . A process called antibody dependent enhancement of infection ( ADE ) , coupled with strong anamnestic cellular immune responses is the leading hypothesis to mechanistically explain more severe disease in secondary infections [6]–[9] . Severe dengue can also occur in primary infection of infants born to dengue-immune mothers with , indicating anamnestic immune responses are not absolutely critical for eliciting the capillary permeability syndrome in all patients . Viral traits may also be important in pathogenesis , with strong evidence that some viral genotypes are fitter than others [10] , [11] . The literature describing the overall relationship between plasma/serum viral burden , disease severity and immune status generally supports the hypothesis that there is a positive correlation between markers of viral burden in the first 2–3 days of fever and the severity of clinical outcomes . For example , during the febrile and early convalescent periods , Taiwanese adults with secondary DENV-3 infections and DHF had higher plasma levels of viral RNA than did patients with DF [12] . An association between higher peak viremia and increased disease severity was observed in Thai children with acute DENV-1 and -2 infections [13] . Similarly , DHF was associated with higher plasma viremia early in illness in Thai children with secondary DENV-3 infections [14] . Duyen et al recently showed that DENV-1 infections were associated with higher viremia and NS1 antigenemia than DENV-2 infections in ambulatory Vietnamese paediatric patients [15] . In the same patients , viremia and NS1 antigenemia persisted for longer in patients with primary infections . In adults , where the risk of clinically apparent disease occurring in primary infection is possibly greater [16] , [17] , there is less evidence relating virological features of infection to immune status or clinical outcome . Kuberski et al reported in 1977 that the magnitude of viremia in young adult patients was higher in primary than secondary DENV-1 infections [18] . More recently , DENV viremia levels from Taiwanese adult patients were reported to be lower in secondary than in primary DENV-2 infections [19] . The dynamics of virus clearance might also be relevant to clinical outcome . In Thai children , Vaughn and others have shown that the slope of the descending portion of the viremia curve was steeper for patients with secondary infection versus those with primary infection and viremia decreased more quickly for patients with DHF than for patients with DF at defervescence [13] . The accelerated clearance of viremia in secondary infection most likely reflects the contribution of anamnestic humoral and cellular immune responses , which themselves have been implicated in the pathogenesis of capillary leakage . Conversely however , Wang et al suggested clearance of the virus and virus-containing immune complexes was slower in adult DHF patients [20] . A better understanding of the relationship between biomarkers of virus infection , the immune response and disease evolution is critical for the rational use of intervention therapies in dengue , e . g . anti-viral drugs or immune-modulating therapies . To this end , this study describes the kinetics of viremia and NS1 in an intensively investigated cohort of Vietnamese adults with dengue and less than 72 hrs of fever enrolled in a randomized placebo-controlled trial ( RCT ) .
A double blind RCT of chloroquine ( CQ ) in 307 adults hospitalized for suspected DENV infection was conducted at the Hospital for Tropical Diseases ( Ho Chi Minh City , Vietnam ) between May 2007 and July 2008 . Information on recruitment , inclusion criteria , randomization , treatment and investigations have been published previously [21] . The Scientific and Ethical committee of the HTD and the Oxford Tropical Research Ethical Committee approved the study protocol and all patients gave written informed consent . The trial was registered with the ISRCTN Register ( ISRCTN38002730 ) . Herein we describe the clinical and virological features of the 257 patients with laboratory confirmed dengue enrolled into this trial , which hitherto have not been described in detail . A diagnosis of laboratory confirmed dengue was reached using serological , antigen detection and molecular methods [22] . In brief , RT-PCR detection of DENV RNA in plasma was performed using an internally controlled , serotype-specific , real-time RT-PCR TaqMan assay that has been described previously [23] . RNA extraction from plasma samples was automated ( NucliSens easyMAG , BioMerieux , Lyon , France ) . Results were expressed as cDNA equivalents per mL of plasma . A capture IgM and IgG ELISA ( MAC and GAC ELISA ) using DENV/JEV antigens and mAb reagents provided by Venture Technologies ( Sarawak , Malaysia ) , was performed as previously described [24] . NS1 was detected by using the NS1 Platelia ELISA assay from BioRad ( Hercules , CA ) according to the manufacturer's instructions . Samples defined as equivocal in the NS1 Platelia ELISA assay were repeated and if they were still equivocal they were regarded as being negative . The interpretation of primary and secondary serological responses was based on the magnitude of IgG ELISA units in early convalescent plasma samples taking into account the illness day . The cut-off in IgG ELISA units for distinguishing primary from secondary dengue by illness day was calibrated using a panel of acute and early convalescent sera from Vietnamese dengue patients that were assayed at the Centre for Vaccine Development , Mahidol University , Bangkok , Thailand using a reference IgM and IgG antibody capture ELISA described previously [25] . Clinical history and examination findings were recorded daily into case record forms . An ultrasound was performed in all patients within 24 hrs of defervescence . Venous blood samples were collected at hospital admission , then twice daily ( around 9am and 3pm ) for a minimum of 5 days after hospital admission and again 10–14 days after discharge from the hospital . A complete blood count , including hematocrit ( Hct ) and platelet measurements , was performed daily for all patients . Hct measurements were performed more frequently if clinically indicated . The extent of hemoconcentration during symptomatic illness was determined by comparing the maximum Hct recorded during hospitalization with either the value recorded at follow-up when available ( 191/248 i . e . 77% of the patients ) or against a sex- and age-matched population value . Plasma was stored frozen in multiple aliquots at −80°C until use in the real-time RT-PCR and NS1 ELISA . The day of fever onset was self-reported by the patient and was designated illness day 1 . DF and DHF were diagnosed according to 1997 World Health Organization ( WHO ) classification criteria and was applied to each case after review of study notes [26] . The 1997 definitions were used for this study because at the time of clinical assessment the 2009 WHO Guidelines and revised classification scheme was not available . DF was defined as a laboratory confirmed dengue case with no evidence of capillary permeability as defined for a DHF case . DHF was defined as laboratory confirmed dengue case with thrombocytopenia ( <100 , 000 platelets/mm3 ) , any hemorrhagic manifestation , and evidence of plasma leakage ( as denoted by a >20% increase in the Hct from the baseline value or by the presence of pleural or abdominal effusions ) . The data used in this analysis was taken from a randomised controlled treatment trial of dengue . Since the intervention ( CQ ) had no measurable impact on virological or immunological outcomes , for the purposes of this analysis we did not distinguish between patients in the CQ or placebo arms of the study . All statistical analysis was performed and figures designed using the software R ( version 2 . 10 . 1 ) . Significance was assigned at P<0 . 05 and were two-sided unless otherwise indicated . Uncertainty was expressed by 95% confidence intervals . The Kruskal-Wallis rank sum test was used for continuous variables and the Fisher's exact test for categorical variables . For the viremia kinetics analysis , when the RT-PCR signal was below the assay limit of detection ( defined as the last dilution of standard that gave a specific signal ) , a value equal to concentration of the last dilution of standard that gave a specific signal divided by 10 was assigned . The maximum viremia level was defined as the highest plasma viremia level measured during illness . The maximum viremia level was considered to be a peak viremia level only in cases in which viremia rose after the enrolment specimen . To compare kinetics of viremia between patients with different serological status , disease severity and serotype , the means of log-transformed viremia measurements made on the same illness day were used as a summary measure of the viremia on that day . To estimate the maximum daily rate of DENV clearance , the slope of the viremia curve was calculated for each illness day as the change in the means of log-transformed viremia measurements made on the same illness day . Only the maximum decreasing daily rate of each patient was used for analysis . Survival analysis using the Kaplan-Meier method and log-rank test was used for all time-to-event outcomes . Time to resolution of viremia or NS1 antigenemia was defined as the time from the start of symptoms until the first of two consecutive plasma samples below the RT-PCR limit of detection or NS1 ELISA negative . The fever clearance time ( FCT ) was defined as the time from the start of symptoms to the start of the first 48 hours period during which axillary temperature remained below 37 . 5°C .
Of the 307 adults with suspected dengue enrolled in the CQ RCT between May 2007 and July 2008 , 257 had laboratory-confirmed dengue including 248 patients with a defined serological and clinical classification and 9 patients with ambiguous or unknown clinical outcomes or serology ( mainly because they left the study prematurely ) . The characteristics of the study population are summarized in Table 1 ( and Table S1 ) . DENV-1 ( 57 . 3% ) was the commonest serotype detected in this population of patients , then DENV-2 ( 20 . 6% ) , DENV-3 ( 15 . 7% ) and DENV-4 ( 2 . 8% ) . DHF was significantly associated with secondary infection compared with primary infection ( 65/215 vs 4/33 i . e . 30 . 2% vs 12 . 1% , P = 0 . 04 , Odds ratio ( OR ) = 3 . 13 , 95% CI 1 . 04–12 . 75 ) . DHF resulting from secondary infection was more commonly associated with DENV-2 ( 21/45 ( 46 . 7% ) ) than for other serotypes ( DENV-1: 33/124 ( 26 . 6% ) , DENV-3: 10/33 ( 30 . 3% ) and DENV-4: 1/7 ( 14 . 3% ) ( DENV-2 vs DENV-1 , -3 and -4 P = 0 . 02 , OR = 2 . 38 , 95% CI 1 . 14–4 . 96 ) ( Table S1 ) . Median viremia levels by illness day for DENV-1 , -2 and -3 are shown in Figure 1 ( and Table S2 ) . In DF patients with primary infection , DENV-1 viremia levels were significantly higher than DENV-2 or DENV-3 levels at multiple time-points during the acute illness ( Figure 1A ) . In DF patients with secondary infection , the most common serological and clinical grouping , and DHF patients with secondary infection , DENV-1 levels were significantly higher than DENV-2 levels and there was also a non-significant trend towards higher DENV-1 levels than DENV-3 levels ( Figure 1B and C ) . Collectively , and despite small sample sizes for some subgroups , these data suggest that DENV-1 infections were associated with higher viremias ( as measured by qRT-PCR ) than DENV-2 , irrespective of disease severity and immune status . DENV-1 was the commonest serotype detected in this patient population and therefore there was sufficient data to enable direct comparisons of viremia kinetics across serological states and clinical severity whilst controlling for the infecting serotype ( Table S2 ) . These data show that in the early acute phase ( illness day 3 ) patients with DENV-1 infection and DHF had significantly higher viremia levels than DENV-1 patients with DF , irrespective of the patient immune status ( Figure 2 ) . These data show also that later in the acute phase ( from day 4 of illness ) patients with primary DENV-1 infections had significantly higher viremia levels than patients with secondary DENV-1 infections , irrespective of the disease severity ( Figure 2 ) . A limitation of these analyses is that in the majority of patients with secondary infections the viremia was already declining at the time of enrolment i . e . we did not observe an obvious peak viremia ( Table S3 ) . Overall , a peak viremia was significantly less often observed in secondary infections than in primary infections for all disease severity grades ( P = 0 . 001 , OR = 3 . 64 , 95% CI 1 . 55–8 . 74 ) . However , there were no significant differences in the duration of illness prior to enrolment between patients in different categories of serological status or disease severity ( P between 0 . 11 and 0 . 96 if all the serotypes are considered , and 0 . 16 and 0 . 39 if and only DENV-1 ) , suggesting this snapshot of viremia levels is unbiased by differences in duration of illness at study enrolment . Peak viremia levels were identified in 72 patients . Peak viremia occurred significantly earlier in secondary DF than in primary DF ( P = 0 . 008 ) and in secondary DHF than in primary DHF ( P = 0 . 04 ) but there were no significant differences between primary DF and primary DHF ( P = 0 . 73 ) and between secondary DF and secondary DHF ( P = 0 . 13 ) for the peak viremia time ( Table S3 ) . Amongst DENV-1 infected patients , peak viremia levels , when observed ( in 51 of 142 DENV-1 infected patients ) , happened significantly earlier in secondary DF than in primary DF ( P = 0 . 0006 ) but , possibly because of small sample size , not in secondary DHF compared to primary DHF ( P = 0 . 31 ) . There was no significant difference between secondary DF and secondary DHF ( p = 0 . 48 ) but there was a non-significant trend towards later viremia peaks in primary DF than in primary DHF ( P = 0 . 052 ) . There were sufficient observations of the magnitude of peak viremia in DENV-1 infections to look for associations with clinical outcome in this subgroup ( Table S3 ) . Peaks of viremia were observed in 9 primary DF , 29 secondary DF , 3 primary DHF and 10 secondary DHF DENV-1 infected patients . Peak viremia levels were not significantly different between DF and DHF patients ( DF vs DHF log10 median peak levels 9 . 89 vs 10 . 27 , P = 0 . 28 ) but there was a non-significant trend towards higher peak viremia during secondary infections than primary infections ( primary vs secondary P = 0 . 096 and primary DF vs secondary DF P = 0 . 086 ) . If considering the highest viremia levels ( as distinct from peak viremia levels ) in DENV-1 patients , these were significantly higher in DHF than in DF ( log10 median levels 9 . 84 vs 9 . 19 , P = 0 . 03 ) . Because most DHF cases were associated with secondary infections , for which peak viremia had already past by the time of enrolment , this difference is probably underestimated . These results suggest secondary infections are generally associated with earlier peak viremia but do not provide any conclusive evidence of higher peak viremia levels in DHF and/or secondary infections . In the 239 patients with detectable viremia , the median of maximum daily rates of DENV clearance ( estimated as the slope of the steepest descending daily portion of the viremia curve ) was 2 . 2 log10 per day in primary DF , 2 . 8 log10 per day in secondary DF , 2 . 1 log10 per day in primary DHF and 3 . 0 log10 per day in secondary DHF . The maximum daily rate of clearance was significantly higher in patients with secondary infections ( median of the maximum daily loss 2 . 9 logs per day ) versus those who experienced primary infections ( median of maximum daily losses = 2 . 1 logs per day , P<0 . 00001 ) ( primary DF vs secondary DF P = 0 . 00004 and primary DHF vs secondary DHF P = 0 . 025 ) . The results were very similar when considering only DENV-1 patients for analysis ( data not shown ) . These data suggest secondary infection is associated with steeper declines in viremia . Amongst all viremic patients ( n = 239 ) time to resolution of viremia was significantly longer in primary infections than in secondary infections ( hazard ratio ( HR ) = 2 . 88 , 95% CI 1 . 79–4 . 63 , log rank test P = 0 . 000005 ) , in primary DF than in secondary DF ( HR = 2 . 60 , 95% CI 1 . 57–4 . 32 , log rank test P = 0 . 0001 ) and in primary DHF than in secondary DHF ( HR = 4 . 92 , 95% CI 1 . 19–20 . 32 , log rank test P = 0 . 015 ) ( Figure 3A ) . Median times to resolution of dengue viremia were 148 hrs ( IQR 140–173 hrs ) in primary DF , 162 hrs ( 134–>171 hrs ) in primary DHF , 120 hrs ( 97–141 . 5 hrs ) hrs in secondary DF and 123 hrs ( 113–138 hrs ) in secondary DHF . Amongst DENV-1 infected patients only ( n = 142 ) , times to resolution of viremia were also significantly longer in primary infections than in secondary infections ( HR = 4 . 21 , 95% CI 2 . 12–8 . 35 , log rank test P = 0 . 000009 ) , in primary DF than in secondary DF ( HR = 3 . 67 , 95% CI 1 . 76–7 . 62 , log rank test P = 0 . 0002 ) and in primary DHF than in secondary DHF ( HR = 7 . 00 , 95% CI 0 . 94–52 . 17 , log rank test P = 0 . 03 ) ( Figure 3B ) . Median times to resolution of DENV-1 viremia were 162 hrs ( IQR 144–>176 hrs ) in primary DF , >171 . 1 hrs ( 134–>179 hrs ) in primary DHF ( since less than 50% of primary DHF had cleared viremia before discharge ) , 125 hrs ( 99–150 hrs ) in secondary DF and 127 hrs ( 107–143 . 5 ) in secondary DHF . Of the 248 patients with defined serological and clinical classifications , there were 214 patients NS1 positive at the time of study enrolment ( plus 2 patients negative at enrolment but NS1 positive 24 and 42 hrs later ) . Consistent with the viremia findings , times to resolution of NS1 antigenemia were significantly longer in primary infections than in secondary infections ( HR = 4 . 57 , 95% CI 2 . 01–10 . 40 , log rank test P = 0 . 00007 ) , in primary DF than in secondary DF ( HR = 3 . 66 , 95% CI 1 . 47–9 . 07 , log rank test P = 0 . 003 ) , in primary DHF than in secondary DHF ( HR = 7 . 04 , 95% CI 0 . 97–51 . 15 , log rank test P = 0 . 02 ) but also in secondary DF than in secondary DHF ( HR = 1 . 86 , 95% CI 1 . 29–2 . 67 , log rank test P = 0 . 0007 ) ( Figure 4A ) . Interestingly , only 5 of 25 patients ( i . e . 20% ) with primary DF and 1 of 4 patients ( i . e . 25% ) with primary DHF had cleared NS1 when discharged from hospital . In patients with secondary dengue , 69 of 127 with secondary DF ( i . e . 54 . 3% ) and 51 of 60 with secondary DHF ( i . e . 85% ) had cleared NS1 when discharged from hospital . Median times to resolution of NS1 antigenemia since illness onset were >166 hrs ( >146–>178 hrs ) in primary DF and >158 hrs ( 138–>171 hrs ) in primary DHF ( since less than 50% of primary DF and primary DHF had cleared NS1 before discharge ) , and 137 hrs ( 105–>174 hrs ) in secondary DF and 121 hrs ( 100–153 hrs ) in secondary DHF . Amongst DENV-1 infected patients ( n = 142 ) , 134 were NS1 positive at the time of study enrolment . Times to resolution of NS1 antigenemia were significantly longer in DENV-1 primary infections than in DENV-1 secondary infections ( HR = not applicable , log rank test P = 0 . 00008 ) , in DENV-1 primary DF than in DENV-1 secondary DF ( HR = 3 . 66 , 95% CI 1 . 47–9 . 07 , log rank test P = 0 . 003 ) , in DENV-1 primary DHF than in DENV-1 secondary DHF ( HR = 7 . 04 , 95% CI 0 . 97–51 . 15 , log rank test P = 0 . 004 ) and in DENV-1 secondary DF than in DENV-1 secondary DHF ( HR = 1 . 86 , 95% CI 1 . 29–2 . 67 , log rank test P = 0 . 00001 ) ( Figure 4B ) . Strikingly , none of the DENV-1 infected patients with primary DF ( n = 15 ) or primary DHF ( n = 3 ) had cleared NS1 when they were discharged from hospital . In contrast , 36 of 84 ( 42 . 9% ) secondary DF and 29 of 31 ( 93 . 5% ) secondary DHF patients had cleared NS1 when they were discharged . Median times to resolution of NS1 antigenemia since illness onset were >172 . 5 hrs in primary DF , >171 hrs in primary DHF , >174 hrs in secondary DF ( since less than 50% of primary and secondary DF , and primary DHF had cleared NS1 before discharge ) and 121 hrs ( 103–144 hrs ) in secondary DHF . Collectively , these results suggest that DENV infection is cleared earlier and faster in secondary infections than in primary infections . There were 240 patients febrile at enrolment ( plus 2 afebrile patients who developed fever soon after ) . Overall , FCT were significantly longer in primary infections than in secondary infections ( log rank test P = 0 . 037 ) but there was no significant difference between primary DF and secondary DF ( HR = 1 . 44 , 95% CI 0 . 93–2 . 23 , log rank test P = 0 . 096 ) , and between primary DHF and secondary DHF ( HR = 1 . 85 , 95% CI 0 . 67–5 . 14 , log rank test P = 0 . 23 ) ( Figure 5 ) . Median FCT since illness onset was 131 hrs ( IQR 95 . 5–151 . 4 hrs ) in primary DF , 141 hrs ( 135 . 5–160 . 5 hrs ) in primary DHF , 118 hrs ( 93 . 1–140 . 2 hrs ) in secondary DF and 120 hrs ( 105–142 hrs ) in secondary DHF . Consistent with the viremia and NS1 findings , these data indicate primary infection was associated with a longer-lived febrile period . For descriptive purposes , the evolution over time of DENV-1 viremia in the context of white blood cell and platelet counts and percentage hemoconcentration was plotted ( Figure 6 ) and summarised in Table S4 . The highest levels of hemoconcentration and the lowest platelet counts occur when the viremia is close to resolution and when the patient is already or very nearly afebrile . The time to platelet nadir and maximum hemoconcentration was shorter in secondary DENV-1 infections ( P<0 . 01 ) . The data also suggest that leucopenia lasts longer in primary infections than in secondary infections .
The interplay between DEN virus infection and host immune status is postulated to play a central role in the pathophysiology of severe dengue . In this current study , we observed important features of this dynamic . First , early viremia levels were higher in patients with DHF , even if the peak viremia level was often not observed because it occurred prior to enrolment . Second , DENV-1 infections manifested as higher and longer-lived viremias , suggesting serotype dependent differences in infection kinetics . Third , the clearance of DENV viremia and NS1 antigenemia occurs earlier and faster in patients with secondary dengue and is also consistent with a faster time to defervescence . Our findings are in agreement with previous studies that found higher viremias associated with more severe disease [12]–[14] . Our data also suggests that quantitative differences exist between DENV serotypes with respect to the kinetics of viremia and NS1 antigenemia . In particular , DENV-1 infections were associated with higher and frequently longer-lived viremia levels than infections with either DENV-2 or DENV-3 . This is in agreement with recent observations in Vietnamese children and adults [15] , [27] . DENV-2 was associated with secondary infection and severe disease in our study; this is also in accordance with previous studies [2] , [27] , [28] . The mechanisms that facilitate relatively higher viremia in DENV-1 infections relative to DENV-2 infections in our study population , and also recently in Vietnamese children [15] , are unknown . Plausibly , DENV-1 has an intrinsically faster rate of replication in this patient population and thereby attains a higher infected cell mass in vivo than DENV-2 . Clearly , further studies will be needed to explore this . The duration of NS1 antigenemia was shorter in patients with secondary infections and this is consistent with previous studies that have suggested reduced sensitivity of NS1-based diagnostic tests in patients with secondary infections [22] , [29]–[32] . One explanation is that NS1 is less likely to be available for detection when a sufficient level of DENV-reactive IgG ( including anti-NS1 IgG ) develops during secondary infections . This may serve to mask the antigen from detection in the immunoassay , and/or result in rapid clearance of NS1 in the form of immune-complexes . Secondary dengue is associated with faster resolution of viremia infection and shorter duration of fever . Interestingly , the daily rates of virus clearance observed in our study were very compatible with those found by Vaughn et al in Thai children [13] . The early adaptive immune response during secondary infection is dominated by populations of memory B and T cells ( and possibly memory-like NK cells [33] ) and at least some components of this response mediate a strong anti-viral action , as evidence by faster clearance rates of the viremia and the NS1 antigenemia . Clearly however , aspects of this rapid host immune response are clinically deleterious given the epidemiological association between secondary dengue and more severe outcomes , and also the timing of when clinical manifestations of capillary permeability occur . This poses the intriguing question of whether modulating the host immune response ( e . g . through early corticosteroid therapy ) could achieve both a more gradual clearance of the virus and a host immune response that elicits less pathology . Assuming blood viremia is a reasonable surrogate of the whole-body virus burden , then the rapid decline of viremia 48–72 hrs after illness onset , especially in secondary infections that carry higher risk for severe outcomes , has implications for rational design of therapeutic pharmacological interventions . An efficacious anti-viral will need favourable pharmacokinetic properties and potency if it is to impact on the viral burden in a rapid and clinically significant manner . Pharmacological targeting ( e . g . with corticosteroids ) of the host immune response , which accounts for the rapid decline of viremia but which also likely contributes to the capillary permeability syndrome , may equally need to be administered early on in illness e . g . in a prophylactic way , to prevent clinical complications such as DSS . The rapid decline in viremia in secondary infections also highlights the importance of early diagnosis , since early diagnosis will provide the greatest opportunity for an intervention ( e . g . an anti-viral ) , to have an impact . Point of care NS1 diagnostics are available but more can be done to improve their sensitivity [34] . More clinical research is also needed to understand if the sensitivity and specificity of early clinical diagnoses ( and prognosis ) can be improved , particularly in primary health care settings . Recent literature suggests this is feasible [35] , [36] . Current animal models of DENV infection are able to provide for in vivo measurements of anti-viral activity [37] , [38] . However these models do not reproduce the temporal changes in virological biomarkers and clinical manifestations seen in naturally infected dengue patients . The lack of concordance between virological and clinical events in small animal models , and what occurs in patients , needs to be carefully considered when evaluating candidate anti-viral drugs for dengue . For instance , the onset of vascular permeability does not follow the disappearance of virus during enhanced DENV infection in mouse models [38] , [39] . There are several limitations to our study . Our results are derived from hospitalized patients who are not necessarily representative of patients being seen at the primary health care level . Interestingly however , the same themes identified in this study in hospitalized Vietnamese adults were also observed in Vietnamese children presenting to primary health care level clinics in Ho Chi Minh City [15] . Of the 248 patients in this study with a defined serological and clinical classification , 126 received a 3 day course of CQ and , whilst no virological or immunological effects were detected , an effect of CQ on the clinical phenotype cannot be excluded because vomiting was more frequent in this treatment arm , possibly leading to more dehydration [21] . The majority of the patients in this study were infected by DENV-1 . Very few had primary DHF ( 4 of 248 ) . RT-PCR measurements of viremia in plasma may not be an entirely accurate surrogate of the infected cell mass in vivo , although it is certainly a better surrogate than NS1 antigenemia , which persists well after the febrile period and is heavily influenced by the immune status of the host ( i . e . primary versus secondary ) . Viremia measurements assessed by RT-PCR encompass both infectious and non-infectious viral particles , and the relative proportions may vary between the different serological responses and/or serotypes [40] . Assessment of plasma virus titers based on plaque titration would have been a valid alternative approach to measuring virus concentrations in plasma , however this biological assay is more difficult to standardize and validate relative to a RT-PCR assay . Another general limitation of measuring virus concentrations in plasma is that viruses might be sequestered in other tissues but inaccessible to measurement while still playing a role in disease pathogenesis . Our study emphasizes the importance of the period before and just after the onset of fever . This and other studies have established that early viremia levels are associated with disease severity , although they are very clearly not the only determinant of outcome . Very little is known of the virological events in the hours preceding and shortly after fever onset , mainly because this is very difficult to investigate without a good experimental model . An interesting insight was provided by clinical trials of DENV-1 , -3 and -4 monovalent live attenuated vaccines in the 1980s [41]–[43] . These vaccines were not sufficiently attenuated and some volunteers developed dengue fever . These studies suggest that viremic period starts several days before the onset of symptoms This presymptomatic viremic period should not be underestimated because of its possible contribution to DENV transmission to uninfected mosquitoes . In our study , we also observed prolonged times of virus clearance in primary dengue , and long-lived higher viremia levels in DENV-1 infections . These might lead to a higher possibility of human to mosquito virus transmission by maintaining viremia above the infectious level over a longer period of time . Collectively , our findings reveal important patterns of the viremia and NS1 antigenemia kinetics according to the patient immune response , disease severity and virus serotype , and may help for the rational design of clinical trials of therapeutic interventions , especially antivirals . | Dengue is an acute viral disease that affects tens of millions of people annually in tropical and sub-tropical countries . In some cases , this infection happens to be severe and even life threatening . Severe cases have been associated with higher levels of virus in the blood . Several hypotheses have been proposed to explain the occurrence of these cases notably by involving the patient's history of previous DEN virus infection ( s ) . Little is known about the relationships between the evolution over time of virus levels in the blood , the clinical outcome and the previous infection ( s ) history—a better understanding of these features could help in anti-viral drug development . To analyze these relationships , we studied well characterized patients who participated in a clinical trial . The majority of these patients were infected by DENV-1 serotype and had higher levels of virus than those infected by DENV-2 and sometimes DENV-3 serotypes . We also found that patients with more severe symptoms had higher levels of virus in the first days of their illness . We found as well that the virus was cleared faster and earlier from the blood of patients previously infected . These findings are of major importance for further anti-viral drug testing . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"clinical",
"research",
"design",
"virology",
"clinical",
"trials",
"dengue",
"fever",
"neglected",
"tropical",
"diseases",
"dengue",
"biology",
"microbiology",
"viral",
"diseases"
] | 2011 | Kinetics of Viremia and NS1 Antigenemia Are Shaped by Immune Status and Virus Serotype in Adults with Dengue |
In nature , larvae of the fruitfly Drosophila melanogaster are commonly infected by parasitoid wasps , and so have evolved a robust immune response to counter wasp infection . In this response , fly immune cells form a multilayered capsule surrounding the wasp egg , leading to death of the parasite . Many of the molecular mechanisms underlying this encapsulation response are conserved with human immune responses . Our findings suggest that protein N-glycosylation , a common protein post-translational modification of human immune proteins , may be one such conserved mechanism . We found that membrane proteins on Drosophila immune cells are N-glycosylated in a temporally specific manner following wasp infection . Furthermore we have identified mutations in eight genes encoding enzymes of the N-glycosylation pathway that decrease fly resistance to wasp infection . More specifically , loss of protein N-glycosylation in immune cells following wasp infection led to the formation of defective capsules , which disintegrated over time and were thereby unsuccessful at preventing wasp development . Interestingly , we also found that one species of Drosophila parasitoid wasp , Leptopilina victoriae , targets protein N-glycosylation as part of its virulence mechanism , and that overexpression of an N-glycosylation enzyme could confer resistance against this wasp species to otherwise susceptible flies . Taken together , these findings demonstrate that protein N-glycosylation is a key player in Drosophila cellular encapsulation and suggest that this response may provide a novel model to study conserved roles of protein glycosylation in immunity .
Cellular encapsulation of invading organisms is a vital and conserved aspect of insect immunity [1] , [2] . In this immune response the invader is recognized as foreign and then surrounded by a multilayered capsule of immune cells which serves to kill or sequester it . Capsule formation can be targeted against any foreign object too large to be phagocytosed , including parasitoid wasp eggs and wasp larvae , fungal and protozoan parasites , and even abiotic objects [1] . Cellular encapsulation represents an important aspect of resistance against pathogens in insect vectors of human disease [3] , [4] and in the genetic model organism Drosophila melanogaster [5] , [6] and is functionally similar to granuloma formation in vertebrates [7] . In nature , Drosophila are commonly parasitized by a wide range of parasitoid wasps [8] . Among these are larval parasitoids , which attack fly larvae and inject their eggs and venom directly into the larval hemocoel . While flies mount a robust cellular encapsulation response against parasitoid eggs , the wasps' venom contains factors that target host immune mechanisms and allow wasps to evade or suppress capsule formation of their natural hosts [9] , [10] . In Drosophila , cellular encapsulation is accompanied by capsule melanization , and is mediated by immune cells known as hemocytes . Flies have three subtypes of hemocytes: macrophage-like plasmatocytes , which act as immune sentinels and comprise ∼95% of all hemocytes in unattacked larvae; crystal cells , which contain the melanization machinery and comprise ∼5% of all hemocytes in unattacked larvae; and lamellocytes , large flattened cells which function in encapsulation and are induced following parasitoid attack and during pupation [11] , [12] . Egg encapsulation begins when plasmatocytes recognize the wasp egg as foreign and adhere to it . Newly differentiated lamellocytes then bind to this plasmatocyte layer . Subsequent lamellocytes adhere to each other to form a consolidated multilayered capsule around the melanized inner layers , leading to death of the wasp egg [1] , [13] . Genetic and transcriptional analyses have begun to provide insight into the regulation of the encapsulation response [14] , [15] , [16] , [17] , [18] . Despite this progress , the molecular mechanisms underlying the process of encapsulation are still largely unexplored . Cellular encapsulation has long been studied in Drosophila melanotic tumor ( tu ) mutants , a class of mutants that are characterized by precocious lamellocyte differentiation and the encapsulation of self tissues [19] , [20] , [21] . Studies of various tu mutants have revealed the specific binding of the lectin wheat germ agglutinin ( WGA ) to lamellocyte surfaces during conditions promoting self-encapsulation [22] , [23] . WGA recognizes N-acetylglucosamine ( GlcNAc ) residues in protein-linked glycans [24] . The WGA binding observed in tu mutants suggests that protein glycosylation of lamellocyte membrane proteins is increased during the encapsulation response , although a functional role of protein glycosylation in encapsulation has never been demonstrated . Protein glycosylation is one of the most abundant post-translational modifications and plays a variety of roles in protein regulation and recognition [25] . There are two main classes of protein glycosylation , Asn- or N-linked glycans , and Ser-/Thr- or O-linked glycans . WGA has been shown to recognize N-glycosylated proteins found in the plasma membrane or secreted into the extracellular space [26] , [27] , [28] , and O-GlcNAc glycans which are exclusively found intracellularly [27] , [29] . The cell surface WGA staining observed in tu mutants therefore likely reflects the presence of N-glycosylated proteins . The N-linked glycosylation pathway [reviewed in 28] begins with the biosynthesis of a dolichol-linked precursor molecule which is then transferred to an Asn residue in the target protein by oligosaccharyltransferase ( Ost ) . The precursor molecule is then trimmed by the activity of the alpha-glucosidase enzymes to produce oligomannose-type N-glycans . These structures can be further modified by alpha-mannosidase-I and β1 , 2-N-acetylglucosaminyltransferase I ( GnTI ) . GnTI activity initiates the synthesis of hybrid , paucimannose and complex-type N-glycans by adding an N-acetylglucosamine residue to the glycan core . This produces a hybrid-type N-glycan , which can be further modified by a variety of enzymes . Protein N-glycosylation has been biochemically characterized in Drosophila at multiple life stages and in a variety of glycosylation defective mutant backgrounds [30] , [31] , [32] , [33] , [34] , [35] , [36] , [37] . Additional genetic characterization has revealed roles for N-glycosylation in tissue morphogenesis , nervous system development , adult locomotion , oxidative stress resistance , and lifespan regulation [31] , [38] , [39] , [40] . Based on the appearance of WGA positive glycosylated proteins on encapsulating lamellocytes , we hypothesized that the N-linked protein glycosylation pathway was also required for wasp resistance in Drosophila .
To assay glycosylation state , we stained hemocytes from attacked and unattacked larvae with fluorescein isothiocyanate ( FITC ) labeled WGA . In accordance with previous findings [22] , we found that hemocytes from unattacked larvae were negative for WGA staining ( WGA-; Figure 1A , B ) . We then assayed WGA staining following attack by Leptopilina clavipes , a parasitoid wasp from the family Figitidae that provokes an immune response in attacked larvae and is avirulent with respect to D . melanogaster [8] . Hemocytes were stained and observed at the following three time points: 0–24 hours , during which time plasmatocytes surrounded the wasp egg and lamellocyte differentiation began; 24–48 hours when lamellocyte-mediated encapsulation and capsule melanization occurred; and at 48–72 hours when melanotic encapsulation was complete . We found that lamellocytes dissected 0–24 hours post-attack ( PA ) showed low-level WGA staining ( Figure 1C , D ) , while plasmatocytes were still WGA− . In contrast , lamellocytes dissected at 24–48 hours PA showed a strong , speckled WGA staining pattern ( WGA+; Figure 1E , F ) , reminiscent of that seen in tu mutant larvae [22] , [23] . Additionally , WGA+ plasmatocytes were occasionally observed at this time point ( Figure 1G , H ) . Hemocytes dissected from larvae following the completion of melanotic encapsulation ( 48–72 hours PA ) were WGA− ( Figure 1I , J ) . Therefore , the WGA+ phenotype was specific for lamellocytes during the period of encapsulation . These results paralleled previous studies of tu ( 1 ) Sz , a conditional self-encapsulating tu mutant strain . In tu ( 1 ) Sz , WGA+ lamellocytes were found in larvae that were actively encapsulating either self or transplanted foreign tissues , but not in larvae raised in non-encapsulating conditions [22] . In combination , these findings suggest that WGA+ lamellocytes are exclusively found in larvae undergoing active encapsulation . To determine whether this WGA+ phenotype reflects the presence of N-glycosylated proteins , we analyzed the WGA staining pattern of lamellocytes from mutant larvae with demonstrated defects in N-glycosylation . A null allele of Mgat1 , the locus that encodes the homolog of the GnTI enzyme [41] , has been demonstrated to induce altered patterns of N-glycosylation [31] . We stained hemocytes from L . clavipes attacked Mgat11 mutant larvae at 24–48 hours PA . Hemocyte subtypes were easily distinguishable in Mgat11 mutant larvae on the basis of morphology and size and did not differ from wild type cells ( as described in [11] , data not shown ) . We found that mutant lamellocytes had greatly reduced WGA staining ( Figure 1K , L ) . This demonstrated that the dynamic protein glycosylation revealed by WGA staining of lamellocytes is dependent on function of the N-glycosylation pathway , and therefore suggests that WGA is recognizing N-glycosylated proteins on lamellocytes . To test the hypothesis that protein N-glycosylation is required for the cellular encapsulation response , we assayed wasp egg encapsulation rates in larvae with mutations in predicted N-glycosylation pathway genes [pathway illustrated in Figure 2A , Drosophila homolog predictions from the KEGG database , 42] . We exposed larvae to the avirulent wasp L . clavipes , which induces a strong cellular encapsulation response in D . melanogaster following attack , and scored egg encapsulation rate ( results shown in Figure 2B ) . We found that the wild type background control strains ( w1118 , y , w and y;ry ) successfully encapsulated L . clavipes eggs ( w1118: 100% encapsulated , n = 90 eggs; y , w: 85 . 5% encapsulated , n = 117 eggs; y;ry: 97 . 6% encapsulated , n = 125 eggs ) . We next tested mutations targeting the oligosaccharide transferase ( Ost ) , α-1 , 2-mannose trimming and GnTI steps in the N-glycosylation pathway , along with a variety of mutations in enzymes in downstream N-glycan processing pathways . Mutations in the Δ and γ subunits of Ost ( OstΔ , encoded by CG6370 and Ostγ , encoded by CG7830 ) , which are expected to abolish protein N-glycosylation , led to an almost complete inability to encapsulate wasp eggs ( OstΔ: 0% encapsulated , n = 83 eggs , p<0 . 001 relative to y , w control; Ostγ: 12 . 0% encapsulated , n = 92 eggs , p<0 . 001 relative to y , w control ) . Mutation of CG1518 , one homolog of the STT3 subunit of Ost , had no effect on encapsulation ( 98 . 0% encapsulated , n = 98 eggs , p = 0 . 89 relative to y;ry control ) . This suggests either that the STT3 subunit is not required for encapsulation or that another locus is redundant to CG1518 in this process . Mutations in the α-1 , 2-mannosidase homologs α-Man-I and CG11874 ( referred to here as α-Man-Ib ) displayed an intermediate encapsulation phenotype ( α-Man-I: 32 . 4% encapsulated , n = 102 eggs , p<0 . 001 relative to y;ry control; α-Man-Ib: 26 . 3% encapsulated , n = 114 eggs , p<0 . 001 relative to w1118 control ) . This suggests a degree of redundancy between these genes and possibly a third α-1 , 2-mannosidase homolog ( CG31202 ) , perhaps explaining previous observations [30] . Mgat11 mutant larvae were significantly impaired in their encapsulation ability ( 13 . 4% encapsulated , n = 97 eggs , p<0 . 001 relative to w1118 control ) , suggesting that hybrid , or more highly processed types of N-glycan structures are required . We then assayed encapsulation rates in downstream Mgat1-dependent processing steps , and found that α-Man-II mutants were also impaired in encapsulation ability ( 8 . 9% encapsulated , n = 79 eggs , p<0 . 001 relative to w1118 control ) . Mutants in the additional downstream N-glycan processing enzymes α-Man-IIb ( 55 . 5% encapsulated , n = 137 eggs , p = 0 . 009 relative to y , w control ) and FucT6 ( 77 . 9% encapsulated , n = 86 eggs , p<0 . 001 relative to w1118 control ) displayed only minor decreases in encapsulation rate , suggesting either genetic redundancy with additional , untested α-mannosidase and fucosyl transferase loci for these enzymes , or that their products play minor roles in encapsulation . A mutation in Mgat2 had no effect on wasp egg encapsulation ( 89 . 0% encapsulated , n = 91 eggs , p = 0 . 41 relative to y , w control ) . The protein-acetylglucosaminyltransferases Mgat1 and Mgat2 encode unique orthologs of GnTI and GnTII ( respectively ) and act in a sequential manner to process N-linked glycans . The differential requirement for encapsulation suggests that Mgat1-dependent hybrid N-glycans but not Mgat2-dependent complex N-glycans are important for cellular encapsulation . The impaired immune response due to loss of function mutations in multiple N-glycosylation genes suggests that the protein N-glycosylation revealed by WGA+ lamellocytes is essential for cellular encapsulation of parasitoid wasp eggs . Interestingly , increased WGA staining of hemocytes has been observed in other insects following parasitization [43] , [44] , suggesting that N-glycosylation plays a conserved role in encapsulation-mediated immune responses in insects . Of the alleles tested in this survey , only Mgat11 has been biochemically characterized , and these mutants were found to be deficient in N-glycosylation activity [31] . Therefore , we decided to focus further experiments on Mgat1 in order to fully characterize the role of protein N-glycosylation in the cellular encapsulation process . Surprisingly we found that larvae heterozygous for Mgat11 ( in trans to a CyO-GFP balancer ) were also deficient in encapsulation ( 3 . 2% encapsulated , n = 31 eggs ) . The phenotype seen in these heterozygotes suggests that wasp egg encapsulation may be very sensitive to protein N-glycosylation levels; GnTI enzyme activity is reduced to approximately 40% of wild type activity in Mgat11/+ flies , but they are wild type with respect to all other known Mgat1 mutant phenotypes , including male fertility , locomotor activity , and life span [31] . This Mgat11 heterozygote phenotype could reflect either haploinsufficiency for the Mgat1 locus or a synthetic interaction with the balancer chromosome . To distinguish between these possibilities , we outcrossed Mgat11 mutant males to females of the background control strain w1118 . We found that these Mgat11/+ larvae also showed a profound defect in cellular encapsulation , only 12 . 2% of L . clavipes eggs were encapsulated at 48–72 hours PA ( n = 49 eggs , p<0 . 001 relative to control; Figure 3A ) . The decreased encapsulation rate was independent of the gender of the mutant parent; in the reciprocal cross only 15 . 2% of parasite eggs were encapsulated ( n = 99 eggs , p<0 . 001 relative to control ) . The Mgat11/+ phenotype is statistically indistinguishable from the Mgat11 homozygous phenotype ( p = 0 . 90 ) , suggesting that Mgat1 is a haploinsufficient locus for wasp encapsulation . To avoid the larval lethality and other defects associated with Mgat1 deficiency [31] , we decided to further characterize the role of Mgat1 in cellular encapsulation using Mgat11/+ larvae . To test whether the decreased GnTI function in Mgat11 heterozygous larvae was sufficient to disrupt the appearance of WGA+ lamellocytes , we stained lamellocytes from Mgat11/+ larvae with FITC-WGA . Mgat11/+ lamellocytes showed a decrease in WGA staining ( Figure 3B , C ) . The staining phenotype was intermediate between WGA+ w1118 larvae and WGA− Mgat11 null larvae ( compare with Figure 1E–F and K–L ) . This finding is in accordance with the GnTI activity seen in the respective genotypes [31] . The inability of Mgat11/+ lamellocytes with intermediate WGA staining levels to encapsulate wasp eggs suggests that lamellocyte function is extremely sensitive to glycosylation state . The above data demonstrates a role for Mgat1 in the encapsulation of the figitid wasp L . clavipes . To test whether the requirement for Mgat1 in wasp encapsulation is specific for figitid wasps , or whether it is part of a more general encapsulation mechanism , we exposed w1118 and Mgat11/+ larvae to a second , distantly related , avirulent wasp from the family Braconidae , Aphaereta sp . [8] . Aphaereta eggs were encapsulated by w1118 larvae ( 100% encapsulated , n = 79 eggs; Figure 3A ) . Mgat11/+ larvae were significantly impaired in encapsulation of Aphaereta eggs ( 16 . 2% encapsulated , n = 74 eggs , p<0 . 001 relative to control; Figure 3A ) . This demonstrates that the requirement for Mgat1 is not specific to parasitoids from the family Figitidae , and suggests that it may be part of a more general wasp encapsulation response . To determine whether the impaired encapsulation phenotype seen in Mgat11/+ larvae represented failure of the immune response and reduced fly survival , we exposed w1118 and Mgat11/+ larvae to the previously described avirulent wasps L . clavipes and Aphaereta sp . , allowed the larvae to develop to adulthood , and measured the eclosion rate of adult flies . We found that while w1118 larvae developed into adult flies 100% of the time ( n = 84 larvae; Figure 3D ) , only 34 . 3% of Mgat11/+ larvae developed into adult flies ( n = 67 larvae , p<0 . 001 relative to control; Figure 3D ) , and the remainder eclosed as adult wasps . These data show an important role for Mgat1 in wasp resistance . The cellular encapsulation response is mediated by signaling in both of the main immune tissues in Drosophila , the hemocytes and fat body [15] , [18] , [45] . We used tissue specific RNA interference-mediated knockdown to determine where Mgat1 is required . Using a previously characterized transgenic knockdown strain [40] , we expressed dsRNA directed against Mgat1 in hemocytes using the He-Gal4 driver [46] and in the fat body using the C833 driver [47] . When outcrossed to the w1118 background control strain , both drivers were able to efficiently encapsulate L . clavipes eggs ( He-Gal4: 100% encapsulated , n = 79 eggs; C833: 98 . 8% encapsulated , n = 86 eggs; Figure 4A ) . Knockdown of Mgat1 in the hemocytes led to a significant reduction in encapsulation ability ( 51 . 8% encapsulated , n = 112 eggs , p<0 . 001 relative to control; Figure 4A ) . However , when Mgat1 was knocked down in the fat body , encapsulation ability was unaffected ( 100% encapsulated , n = 81 eggs , p = 0 . 37 relative to control; Figure 4A ) . These findings demonstrate a hemocyte-specific role for Mgat1 in cellular encapsulation . The cellular encapsulation response is characterized by an increase in total hemocyte number and by the differentiation of lamellocytes [6] . Several studies have found that hemocyte number is an important determinant of wasp resistance , both within D . melanogaster and between Drosophila species [48] , [49] , [50] . To test whether loss of Mgat1 led to defects in hematopoiesis , we collected hemolymph samples and measured both the total hemocyte count ( THC ) and number of lamellocytes in mutant and control larvae after 72 hour exposure to L . clavipes . We found that hematopoiesis was not reduced in Mgat11/+ larvae ( Figure 4B ) . In fact , the THC of wasp attacked mutants was significantly higher than that seen in wasp attacked control larvae ( THC: w1118 , 6280 hemocytes per larva , Mgat11/+ , 13280 hemocytes per larva , p = 0 . 010 ) . Furthermore , Mgat11/+ larvae had more circulating lamellocytes following wasp attack ( w1118 , 720 lamellocytes per larva , Mgat11/+ , 1520 lamellocytes per larva , p<0 . 001 ) . These findings suggest that the failure to encapsulate wasp eggs observed in Mgat11/+ mutant larvae is not due to reduced hemocyte numbers , but instead must reflect a requirement for protein N-glycosylation in hemocyte function during capsule formation . While scoring the egg encapsulation rate at 72 hours PA , we observed that whereas w1118 attacked larvae contained encapsulated and melanized eggs ( Figure 5A , indicated by arrowheads ) , 56 . 6% of attacked Mgat11/+ mutant larvae ( n = 99 larvae ) had pieces of melanized tissue floating in the hemocoel ( Figure 5B , indicated by arrows ) . These melanized pieces appeared to be broken capsules and were specific to attacked larvae in which a hatched parasitoid larva was found . This broken capsule phenotype was also seen in 65 . 5% of larvae in which Mgat1 was knocked down by dsRNA expression specifically in hemocytes ( n = 87 larvae , data not shown ) , suggesting that loss of Mgat1 in hemocytes is responsible for the appearance of broken capsules . The phenotype was seen at similar levels in six of the seven N-glycosylation pathway mutants that had a significant effect on encapsulation rate ( Ostγ , α-Man-I , α-Man-Ib , α-Man-II , α-Man-IIb , and FucT6 , data not shown; OstΔ mutants do not have the broken capsule phenotype suggesting that either this mutation effects encapsulation in a different way , or has pleiotropic effects that obscures the broken capsule phenotype ) . These data suggest that most N-glycosylation mutants can initiate capsule formation , but are in some way deficient in successfully encapsulating wasp eggs . The major stages of capsule formation were observed using the hemocyte-subtype specific fluorescent markers eater-GFP to mark plasmatocytes and msn-mCherry to mark lamellocytes [51] . Following egg recognition , plasmatocytes adhere to the wasp egg forming the inner capsule layer [13] . In control larvae this process was completed during the first 24 hours PA ( Figure 6A ) and was not disrupted in Mgat11/+ larvae ( Figure 6B ) . This inner plasmatocyte layer was then covered by several layers of lamellocytes [13] that formed during the 24–48 hour PA time point ( Figure 6C ) . This process was also unaffected in Mgat11/+ larvae ( Figure 6D ) , suggesting that egg recognition and cell migration are independent of protein N-glycosylation . Finally , the lamellocytes adhere to each other , forming a consolidated outer capsule [1] , [13] . By 48–72 hours PA , this consolidated outer cell layer covering over the now melanized inner layer was observed in 100% of control larvae ( n = 20 , arrows , Figure 6E , F ) , and individual cells were no longer apparent . However , in Mgat11/+ larvae the lamellocyte layers did not consolidate and instead were observed as thick , loosely attached layers in which individual lamellocytes were still visible ( Figure 6G–I ) . Closer examination revealed distinct outer layer morphologies , and a lack of consolidation in Mgat11/+ larvae ( Figure 6J , K ) . We hypothesized that wasp larvae were capable of hatching from these loosely formed capsules , which then disintegrated into the small melanized pieces observed . Because defective capsules in Mgat11/+ larvae were melanized , and therefore visible in live larvae , we were able to follow the progression of capsule formation and disintegration in control and Mgat11/+ larvae . At 24–48 hours PA , we found that there was no significant difference in the incidence of capsule formation in the two genotypes; 94 . 8% of attacked w1118 larvae ( n = 231 ) and 79 . 2% of attacked Mgat11/+ larvae ( n = 101 , p = 0 . 09 ) showed evidence of capsule formation . To trace the fate of the capsules , these larvae were dissected at 48–72 hours PA . We found that whereas 98 . 2% of the w1118 capsules remained whole ( n = 219; Figure 7A ) , only 10 . 0% of Mgat11/+ capsules were recovered intact ( n = 80 , p<0 . 001 relative to control; Figure 7A ) , and the remainder of the mutant larvae showed the broken capsule phenotype ( Figure 7B ) and contained a live wasp larva ( Figure 7C ) . Examination of the melanized pieces using the msn-mCherry marker demonstrated that they were composed of lamellocytes as would be expected of a broken capsule ( Figure 7D , E ) . This analysis demonstrates that protein N-glycosylation is required for completion of capsule formation and suggests that melanization alone is not sufficient for parasite-killing , but must be accompanied by lamellocyte consolidation . Since Mgat1-dependent protein N-glycosylation was required for wasp resistance , we hypothesized that wasp venom may target the protein N-glycosylation pathway as a virulence mechanism . We previously observed that larvae attacked by the parasitoid wasp L . victoriae occasionally show a broken capsule phenotype similar to that seen in Mgat1 mutants ( see below ) . We therefore assayed lamellocyte glycosylation in L . victoriae attacked larvae to test whether L . victoriae attack altered the WGA staining pattern . We found that these larvae showed decreased lamellocyte WGA staining ( compare Figure 8A , B to 8C , D ) , similar to that seen in Mgat1 mutants . This suggests that L . victoriae venom either directly targets the N-glycosylation pathway or indirectly disrupts protein glycosylation , perhaps by altering host carbohydrate metabolism or upstream signaling events . If the venom acts directly on glycosylation , then ectopic expression of Mgat1 might be predicted to rescue the WGA staining phenotype . Hemocyte-specific expression of full length Mgat1 [40] was sufficient to restore the lamellocyte WGA+ pattern in L . victoriae attacked larvae ( Figure 8E , F ) , suggesting L . victoriae venom directly affects N-glycosylation . We found that 19 . 5% of L . victoriae attacked larvae ( n = 118 ) showed the broken capsule phenotype ( Figure 9A ) and that the appearance of broken capsules was correlated with the presence of live wasp larvae in the larval hemocoel ( Figure 9C ) . This suggests that , like in Mgat1 mutants , the broken capsule seen in L . victoriae attacked larvae may be indicative of a failed encapsulation response . Since expression of Mgat1 in hemocytes restored WGA staining in L . victoriae attacked larvae , we tested to see whether it would also block the immune suppressive effects of L . victoriae venom . We first verified that Mgat1 overexpression did not impair cellular encapsulation ability ( 97 . 4% of L . clavipes eggs encapsulated , n = 39 eggs ) . We then assayed the ability of control and Mgat1 over-expressing larvae to encapsulate the eggs of three strains of L . victoriae ( LvPhil , LvHaw and LvUnk ) . All three were virulent on control larvae: in both LvPhil and LvHaw attacked larvae , 0% of the eggs were encapsulated ( n = 84 and 91 respectively ) , and in LvUnk attacked larvae , 7 . 4% of the eggs were encapsulated ( n = 95; Figure 9E ) . The broken capsule phenotype was suppressed by hemocyte-specific Mgat1 overexpression ( 0% of attacked larvae had broken capsules , n = 120 , p<0 . 001 relative to control ) , which conferred upon larvae the ability to encapsulate L . victoriae eggs ( Figure 9B ) . Accordingly , we observed that these larvae formed a consolidated lamellocyte layer ( arrowhead , Figure 9D ) . Furthermore , this L . victoriae egg encapsulation happened at high frequency in all three tested strains: in LvPhil attacked larvae , 64 . 3% of the eggs were encapsulated ( n = 98 , p<0 . 001 relative to control ) , in LvHaw attacked larvae , 59 . 2% of the eggs were encapsulated ( n = 98 , p<0 . 001 relative to control ) , and in LvUnk attacked larvae , 64 . 7% of the eggs were encapsulated ( n = 99 , p<0 . 001 relative to control; Figure 9E ) . This suggests that targeting protein N-glycosylation is a conserved virulence mechanism in L . victoriae from diverse populations . The ability of fly larvae with hemocyte specific Mgat1 overexpression to encapsulate L . victoriae eggs suggests that decreasing glycosylation ( as reflected in decreased WGA staining ) may be an important aspect of L . victoriae virulence . Based on our results we hypothesize that L . victoriae venom may contain ‘anti-glycosylation’ factors that either block protein N-glycosylation ( directly or indirectly as described above ) , or de-glycosylate membrane surface proteins . We would further predict that since this venom activity is sensitive to GnTI activity , it likely targets the N-glycosylation pathway at , or downstream of , the level of Mgat1 function . To test the specificity of this mechanism , we next assayed encapsulation rates using the sister species to L . victoriae , L . heterotoma [8] . Neither control nor Mgat1 over-expressing larvae were able to encapsulate L . heterotoma eggs ( control: 0% encapsulation , n = 90 eggs; Mgat1 overexpression: 0% encapsulation , n = 88 eggs; Figure 9E ) . L . heterotoma has previously been demonstrated to induce lysis of Drosophila lamellocytes [52] , presumably removing any need for an ‘anti-glycosylation’ role of its venom . This finding demonstrates that the enhanced encapsulation ability conferred by ectopic Mgat1 expression is not generalized to all wasp parasites and that wasp virulence strategies differ greatly even among closely related species . The findings described in this report clearly demonstrate an important role for protein N-glycosylation in the Drosophila cellular encapsulation response against parasitoid wasp eggs , and specifically in the consolidation of the outer lamellocyte layers of melanotic capsules . The identities and functions of the relevant glycosylated protein ( s ) are still unknown , so further study will be required to uncover the exact molecular mechanisms responsible for lamellocyte consolidation . Furthermore , we have uncovered a parasite virulence strategy that is dependent on prevention of lamellocyte protein N-glycosylation and lamellocyte consolidation . Protein N-glycosylation is also a common post-translational modification in human immune cells [53] and in fact many key players in both innate and adaptive immune responses are glycoproteins [54] . It has been demonstrated that protein N-glycosylation is important for regulating protein stability , and for the activation of numerous immune receptors , including the B- and T-cell receptors , along with cytokine and Toll-like receptors [54] , [55] . During an immune response , N-glycosylated proteins play important roles in pathogen recognition and in mediating cell-cell interactions among cells of the immune system [55] , [56] , perhaps in a manner related to the consolidation of Drosophila lamellocytes . Recent findings have revealed that the Drosophila cellular encapsulation process may serve as a useful model of human immune cell function . It is becoming increasingly apparent that the molecular mechanisms underlying cellular encapsulation in Drosophila are highly homologous to those involved in important human immune functions such as immune cell adhesion , wound healing , thrombosis , and inflammation [16] , [57] , [58] . Based on this mechanistic homology and on a shared requirement for protein N-glycosylation [59] , the cellular encapsulation of parasitoid wasp eggs may serve as a novel model to further examine the conserved roles of N-linked protein glycosylation in immunity .
The following Drosophila melanogaster alleles were used in this study: y , w;CG6370DG02210 ( insertion of a P{wHy} element into exon 1 of the locus [60] ) , y , w;CG7830EY16757 ( insertion of P{EPgy2} into exon of the locus 1 [61] ) , y , CG1518KG03333 ( insertion of P{SUPor-P} into exon 1 of the locus [61] ) , y , α-Man-IKG04725 ( insertion of P{SUPor-P} into intron 1 of the locus [61] ) , w1118; CG11874f07221 ( insertion of PBac{WH} into exon 2 of the locus [61] , [62] ) , w1118; α-Man-IIG4901 ( insertion of P{EP} into exon 1 of the locus [61] , [63] ) , y , w;α-Man-IIbKG05078 ( insertion of P{SUPor-P} into intron 1 of the locus [61] ) , y , w;Mgat2EY07798 ( insertion of P{EPgy2} into exon 1 of the locus [61] ) , w1118 , FucT6e02394 ( insertion of PBac{RB} into exon 4 of the locus [61] , [62] , [64] ) ; all from the Bloomington Drosophila Stock Center ) , and w1118;Mgat11 ( provided by G . Boulianne [31] ) . The w1118 , y , w and y;ry strains served as genetic background controls throughout the study . Overexpression and RNAi knockdown experiments were performed with the Gal-4 drivers He-Gal4 [46] and C833 [47] ( both from the Bloomington Drosophila Stock Center ) , and the previously described Mgat1 reagents UAS-Mgat1 and UAS-Mgat1RNAi ( provided by G . Boulianne [40] ) . msn-mCherry and eater-GFP ( provided by R . Schulz [51] ) were used to mark lamellocytes and plasmatocytes , respectively . In this study we used the figitid wasp species L . clavipes , L . victoriae , and L . heterotoma , and the braconid wasp Aphaereta sp . The L . clavipes strain used in this study was provided by J . van Alphen , and the L . victoriae strain LvUnk was provided by S . Govind . The other L . victoriae strains were collected by the Schlenke lab ( LvHaw in Hawaii and LvPhil in the Philippines ) , as was the strain of Aphaereta sp . ( collected in Atlanta , GA ) . The L . heterotoma strain used has been previously described [18] . The laboratory culture of L . heterotoma is maintained on D . melanogaster , L . victoriae is maintained on D . annanassae and Aphaereta sp . and L . clavipes are both maintained on D . virilis . Adult females of each fly strain were allowed to lay eggs onto molasses medium supplemented with yeast paste in 60 mm Petri dishes . After 96 hours , adult flies were removed and second instar fly larvae were collected for wasp exposure . Forty fly larvae were moved into a 35 mm Petri dish filled with 1 mL of Drosophila medium . Three female wasps were then placed onto the dish and allowed to attack for either 24 or 72 hours depending on the experiment ( described below ) . The 72 hour wasp exposure condition allowed for nearly all larvae to be attacked without significant rates of superparasitism . Attack rates for each wasp species were calculated based on three replicates of w1118 wasp exposures , and were as follows: L . clavipes: 1 . 00±0 . 03 eggs/larva , Aphaereta sp . : 0 . 88±0 . 05 eggs/larva , L . victoriae ( average of all three strains ) : 0 . 99±0 . 06 eggs/larva and Lh14: 1 . 02±0 . 08 eggs/larva . Unattacked controls were treated in parallel . For WGA staining , larvae were exposed for 24 hours and dissected at the indicated times . Larvae were bled onto diagnostic slides ( Tekdon , Inc . ) and hemocytes were allowed to adhere for 5 minutes . Hemocytes were then stained with 100 µg/ml FITC labeled WGA from Vector Laboratories ( FL-1021 ) for 3 minutes and washed three times with Drosophila Ringer's solution as described in [22] . Stained hemocytes were visualized using an Olympus BX51 microscope with a FITC filter and Olympus DP2-BSW software . After a 72 hour wasp exposure , thirty fly larvae from each attack plate were dissected to assay attack rate and encapsulation rate . At the end of the exposure period larvae were scored for the presence of an encapsulated wasp egg or live wasp larva . Experiments were performed in triplicate . After a 72 hour wasp exposure , thirty fly larvae were recovered from each plate and allowed to eclose . The total number of flies and wasps eclosed from the treatments were determined 7 and 14 days following infection , respectively . By these times , all viable flies and wasps emerged . Experiments were performed in triplicate . After a 72 hour wasp exposure , five larvae from each plate were removed , rinsed in Drosophila Ringer's solution and bled into 20 µl of 1 µl PBS containing 0 . 01% phenylthiourea ( PTU ) . Hemocytes were then pipetted into a disposable hemocytometer ( Incyto C-Chip DHC-N01 ) and allowed to adhere for 30 minutes . Hemocytes from each sample were counted from sixteen 0 . 25×0 . 25×0 . 1 mm squares . Counts were then normalized to a per larva value . Experiments were performed in triplicate . For whole larva imaging , larvae were chilled and imaged using a Leica stereo-dissecting scope with a Moticam MIP 2 . 0 and Multi-Focus Pro software . For imaging capsules , larvae were dissected into Drosophila Ringer's solution or PBS containing 0 . 01% PTU ( PTU was found to quench fluorescence and so was only used for brightfield imaging ) . Dissected capsules were visualized using an Olympus BX51 microscope with FITC and TRITC filters with Olympus DP2-BSW software . Figures were compiled using Adobe Photoshop . Larvae were exposed to wasps for 24 hours , and then aged a further 24 hours . At this time point larvae were scored for the presence of a capsule in the hemocoel . Larvae with capsules were selected and aged for a further 24 hours . At this time point the larvae were scored for the presence of an encapsulated wasp egg or broken capsule and live wasp larva . Experiments were performed in triplicate . For analysis of encapsulation/eclosion rates and phenotypic penetrance , mutant strains were compared to wild type background controls by Student's t-test ( Numbers ) using data from independent replicates . | Organisms such as the fruitfly Drosophila melanogaster have long been used as model systems to understand complex aspects of human biology . Work on Drosophila antimicrobial immunity has led to identification of mechanisms underlying human innate immunity , such as the use of Toll-like receptors for recognizing antigen and initiating humoral immune responses . Flies and humans are also infected by larger parasites against which they mount immune blood-cell based responses , but the genetic basis for cellular immunity is poorly characterized . In nature , flies are often infected by parasitoid wasps that lay their eggs in fly larvae , inducing a cellular immune response in the flies . Fly blood cells surround the wasp egg and form a tightly connected capsule leading to death of the egg in a process called encapsulation , which is similar to human granuloma formation . In this study we identified eight new genes that are important for encapsulation . These genes are part of the N-glycosylation pathway , and we found that without N-glycosylation of proteins on blood cell surfaces , capsules surrounding wasp eggs cannot consolidate into a tight capsule , allowing the wasps to escape . Interestingly , we also found a wasp that disrupts N-glycosylation so that it can evade the encapsulation response . Our work may provide a model to better understand the role of N-glycosylation in human immunity . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] | [
"animal",
"models",
"drosophila",
"melanogaster",
"model",
"organisms",
"genetics",
"of",
"the",
"immune",
"system",
"immunity",
"innate",
"immunity",
"genetics",
"parasitology",
"immunology",
"biology",
"zoology",
"genetics",
"and",
"genomics",
"immune",
"response"
] | 2012 | Mgat1-dependent N-glycosylation of Membrane Components Primes Drosophila melanogaster Blood Cells for the Cellular Encapsulation Response |
The major histocompatibility complex ( MHC ) region is strongly associated with multiple sclerosis ( MS ) susceptibility . HLA-DRB1*15:01 has the strongest effect , and several other alleles have been reported at different levels of validation . Using SNP data from genome-wide studies , we imputed and tested classical alleles and amino acid polymorphisms in 8 classical human leukocyte antigen ( HLA ) genes in 5 , 091 cases and 9 , 595 controls . We identified 11 statistically independent effects overall: 6 HLA-DRB1 and one DPB1 alleles in class II , one HLA-A and two B alleles in class I , and one signal in a region spanning from MICB to LST1 . This genomic segment does not contain any HLA class I or II genes and provides robust evidence for the involvement of a non-HLA risk allele within the MHC . Interestingly , this region contains the TNF gene , the cognate ligand of the well-validated TNFRSF1A MS susceptibility gene . The classical HLA effects can be explained to some extent by polymorphic amino acid positions in the peptide-binding grooves . This study dissects the independent effects in the MHC , a critical region for MS susceptibility that harbors multiple risk alleles .
Across the entire human genome , the major histocompatibility complex ( MHC ) on chromosome 6 makes the single largest contribution to multiple sclerosis ( MS ) susceptibility . The classical HLA-DRB1*15:01 allele has been documented as the strongest association to MS risk , and its role has been studied and replicated extensively [1] . Numerous other HLA alleles have been suggested to be associated with MS susceptibility , but the complex structure of the MHC has made it challenging to unequivocally pinpoint variants that play a causal role in MS [1] , [2] . For example , it has been suggested that DQB1*06:02 , an MHC class II allele in strong linkage disequilibrium ( LD ) with DRB1*15:01 , either has no independent effect [3] or acts in an extended haplotype with DRB1*15:01 , the DRB1*15:01—DQB1*06:02 haplotype , or the DRB1*15:01—DQA1*0102—DQB1*06:02 haplotype [4] , [5] . The ambiguity and the lack of replication for many of the MHC associations can be attributed to the extended LD structure of the MHC [6] , the limited number of HLA loci analyzed , and the relatively small sample size of previous studies . Thanks to a large sample size and a novel procedure to impute classical HLA alleles from SNP data , a recent study described independent MHC effects for DRB1*15:01 , *03:01 and *13:03 as well as HLA-A*02:01 and rs9277535 [7] . In the present study we sought to test not only the role of classical HLA alleles but also of potentially functional variation within the HLA genes . To this end , we imputed classical alleles as well as their corresponding amino acid sequences in 8 HLA genes in a large population of 5 , 091 MS cases and 9 , 595 healthy controls , with genome-wide data ( GWAS ) , following a recently described imputation protocol [8] . Both the samples and the imputation method used were independent of recent efforts exploring MHC associations to MS susceptibility [7] .
The most statistically significant variant in the univariate analysis ( see Material and Methods for details ) was HLA-DRB1*15:01 ( odds ratio [OR] = 2 . 92 , p = 1 . 4×10−234 , Figure 1A ) . Looking at each category of variants ( SNPs , two-digit HLA alleles , four-digit HLA alleles and amino acid positions ) , the amino acid position with the smallest p-value was position −5 in the leader peptide of DQβ1 ( p = 7 . 6×10−231 ) , and the most statistically significant SNP was at position 32 , 742 , 280 ( OR for the A allele = 2 . 96 , p = 5 . 1×10−229 ) . An equivalent effect was observed for HLA-DQB1*06:02 ( OR = 2 . 96 , p = 5 . 4×10−229 ) . We first tested whether the DRB1*15:01 effect could be explained by DQB1 . Adjusting for DQB1 variants , we observed that DRB1*15:01 always had a residual effect ( p<10−6 ) . Conversely , adjusting for DRB1*15:01 , the effect of DQB1 variants were accounted for ( p>0 . 8 ) , suggesting that DRB1*15:01 had a non-equivalent and more statistically significant effect than the DQB1 variants . Furthermore , the extended DRB1*15:01—DQB1*06:02 haplotype ( p = 7 . 5×10−231 ) did not improve upon the association of DRB1*15:01 alone . Similarly , the classical DQA1*01:02 allele—that was suggested to contribute to the effect of the haplotype—was strongly associated ( p = 4 . 8×10−178 ) , but its effect could be entirely explained by DRB1*15:01 . These observations strengthen the hypothesis that the primary MHC effect in MS is mediated by DRB1*15:01 and not by variants in the DQB1 or DQA1 loci . The DRB1 locus ( all four-digit alleles in one model ) had a p-value of 4 . 0×10−263 in the initial analysis ( Figure 1B ) . After adjusting for DRB1*15:01 , the residual DRB1 locus effect ( due to all remaining DRB1 four-digit alleles ) was still statistically significant ( p = 3 . 1×10−37 ) , indicating the presence of multiple independent DRB1 effects . Applying a forward stepwise strategy ( see Materials and Methods for details ) , we established statistical independence for 5 additional DRB1 alleles: *03:01 , *13:03 , *04:04 , *04:01 , and *14:01 ( Table S1 ) . After controlling for the effects of all 6 significant DRB1 alleles ( including *15:01 ) , there was no evidence for a residual signal ( p = 1 . 5×10−05 ) . We also applied several other variant selection approaches to test the robustness of these findings; all approaches identified the same six alleles ( Table S1 ) . Having analyzed the effects at HLA-DRB1 , we tested all other variation across the MHC while correcting for the six statistically independent DRB1 alleles , namely DRB1*15:01 , DRB1*03:01 , DRB1*13:03 , DRB1*04:04 , DRB1*04:01 , and DRB1*14:01 . The most statistically significant variant was SNP rs2844821 near HLA-A ( OR for G allele = 0 . 70 , p = 3 . 2×10−29 , Figure 1C ) . Due to LD , this SNP effect is statistically equivalent to the effect of HLA-A*02:01 ( OR = 0 . 70 , p = 7 . 4×10−29 ) and amino acid Val at position 95 in the peptide-binding groove of the HLA-A protein ( OR = 0 . 70 , p = 9 . 6×10−29 , Figure 2 ) . Controlling for this effect , there were no other HLA-A associations . Controlling for the 6 DRB1 alleles and the HLA-A effect , the next most statistically significant variant was rs9277489 ( OR for C = 1 . 31 , p = 2 . 6×10−18 ) . This SNP is in the intronic region of HLA-DPB1 gene and in perfect LD ( r2 = 1 , based on HapMap Phase II ) with rs9277535 that was previously associated with MS susceptibility [7] , [9] . The most statistically significant HLA allele was DPB1*03:01 ( p = 3 . 6×10−15 ) , but the effect of rs9277489 cannot be explained by DPB1*03:01 alone ( p = 1 . 7×10−06 for rs9277489 in the presence of DPB1*03:01 ) . The most statistically significant amino acid mapped to position 65 of HLA-DPβ1 ( OR for Leu vs . Ile = 1 . 37 , p = 3 . 7×10−18 ) , which explained the effect of rs9277489 ( p = 0 . 003 for rs9277489 in the presence of Leu65 in HLA-DPβ1 ) . This amino acid is also located in the peptide-binding groove of HLA-DPβ1 ( Figure 2 ) . After controlling for rs9277489 , there was no residual effect at the DPB1 locus ( p>1 . 0×10−5 ) . Adjusting also for the DPB1 effect , we identified rs2516489 as the next most statistically significant variant ( OR for T = 1 . 31 , p = 6 . 7×10−13 , Figure 1G , Figure S2B ) . This SNP tags a region of extended LD containing several non-classical MHC class I , class III and cytokine genes , i . e . MICB , DDX39B ( BAT1 ) , NFKBIL1 , TNF , LTA , LTB , and LST1 ( Figure 3 ) . We note that this region had no substantial effect in the univariate analysis ( Figure 1A , Figure S2A ) , but it became genome-wide significant once the DRB1*15:01 effect was accounted for ( Table S2 ) . There was no evidence of interaction either with DRB1*1501 ( Table S2 ) or any other of the identified effects . To explore this phenomenon further , we stratified the samples according to the presence of DRB1*15:01 into carriers ( heterozygous and homozygous ) and non-carriers . Univariate analysis in these two strata revealed a consistent but modest effect ( OR ∼1 . 2 ) for the associated SNP in both DRB1*15:01 carriers and non-carriers ( Table S2 , Figure S3 ) . This phenomenon can likely be explained by Simpson's paradox , where two subgroups share the same association but the overall population shows no association ( or even a reversed one ) [10] . This analysis therefore returns , for the first time , robust evidence supporting the role of non-HLA genes within the MHC . To explore any functional consequences of the SNPs in the MICB-LST1 region we tested these SNPs for cis-eQTL ( expression quantitative trait loci ) effects in peripheral blood mononuclear cells ( PBMCs ) of 213 MS subjects [11] as well as CD4+ T cells and CD14+ monocytes of 211 healthy controls ( Table S3 ) . None of the associated SNPs had a strong cis-eQTL effect ( p>1×10−5 ) : the strongest effect in this region is the relation of rs2516489 to LST1 expression ( p = 1 . 91×10−5 ) in the CD4+ T cells of healthy individuals . The next strongest effect also involved rs2516489 but was seen in relation to HCG18 ( p = 3 . 19×10−5 ) in the PBMCs of MS subjects . None of the SNPs had a statistically significant cis-eQTL effect on any of the class I or II classical HLA genes ( Table S3 ) . Leveraging the publicly available Encyclopedia of DNA Elements ( ENCODE ) [12] and NIH Epigenomics Roadmap [13] for immune cells and cell lines it is evident that the region has an abundance of functional elements ( Figure 3 ) . Of specific interest is the non-coding naturally occurring read-through transcription between the neighboring ATP6V1G2 ( ATPase , H+ transporting , lysosomal 13 kDa , V1 subunit G2 ) and DDX39B ( DEAD box polypeptide 39B ) genes . Two SNPs , rs2523512 and rs2251824 , tag this element that has a strong signal in the DNase hypersensitivity assay in all immune cell types , suggesting that it is an active cis-regulatory region . The histone markers for promoters , enhancers and active elongation also support these data , while this region is identified as an active transcription start site using chromatin states [14] . Other candidates are the TNF and LTB genes . Rs2516489 , the SNP with the best ( but not statistically significant ) cis-eQTL effects , lies within a region of heterochromatin , with no indication of regulatory potential in the available data . Adjusting for 6 classical DRB1 alleles , HLA-A*02:01 , rs9277489 ( HLA-DPB1 effect ) and rs2516489 , we observed another novel signal emerging from the HLA-B locus ( p = 7 . 9×10−11 ) . The most statistically significant variants were HLA-B*37 , HLA-B*37:01 , amino acid Ser at position 99 in HLA-B ( Figure 2 ) and a SNP in position 31 , 431 , 006 ( hg18 ) ( Figure 1I , J ) . All of these variants had statistically equivalent effects ( OR = 1 . 75 , p = 2 . 2×10−08 ) . Accounting for the effect of HLA-B*37:01 , no other variant in HLA-B exceeded our a priori defined threshold , although the residual effect at the HLA-B locus due to all remaining classical HLA-B alleles was still statistically significant in our analysis ( p = 6 . 5×10−06 , Figure 1L ) . This residual association could be accounted for by HLA-B*38:01 ( OR = 0 . 55; p = 4 . 1×10−05 ) . After adding HLA-B*38:01 to the model , there was no longer evidence for a residual effect of classical HLA-B alleles ( p>0 . 002 ) or elsewhere across the MHC . No amino acid position in HLA-B could explain the HLA-B*38:01 effect . Next , we set out to assess whether a specific set of amino acids within the HLA-DR molecule could explain the collective effect of the six classical DRB1 alleles identified above . To this end , we tested each polymorphic amino acid position using an omnibus test ( a regression model with all but one amino acids carried by a given position ) , adding all amino acids ( but one ) of the most statistically significant position to the model in a forward stepwise fashion . The most significant amino acid position in DRβ1 mapped to position 71 ( p = 1 . 2×10−227 , Figure S4A ) , which carries 4 possible alleles: Ala , Arg , Glu , and Lys . Controlling for the alleles at position 71 ( df = 3 ) , there was still a strong residual signal for DRB1*15:01 ( p = 5 . 8×10−13 ) , indicating that amino acid position 71 alone does not explain the DRB1*15:01 effect . Adjusting for the alleles at position 71 , position 74 was the next most statistically significant ( p = 1 . 2×10−16 , Figure S4B ) . This position harbors five possible alleles: Arg , Leu , Glu , Ala and Gln . Controlling for positions 71 and 74 , position 57 ( with four alleles: Asp , Ser , Val or Ala ) was the next most statistically significant association ( p = 4 . 9×10−11 , Figure S4C ) . Controlling for positions 71 , 74 and 57 , we found position 86 as the most statistically significant association ( OR = 1 . 35 for Val vs . Gly , p = 1 . 0×10−06 , Figure S4D ) . After controlling for these four positions , no other amino acid position exceeded our significance threshold ( Figure S4E ) , although HLA-DRB1*15:01 still showed a residual association signal ( p = 10−05 ) . The model with the four DRβ1 amino acid positions could explain the data better than a model with only DRB1*15:01 ( p = 2 . 6×10−26 in favor of the DRβ1 amino acid positions ) , but it was slightly worse than the model with the six DRB1 alleles ( p = 0 . 001 in favor of the 6 DRB1 alleles ) . All four amino acid positions reside in the peptide-binding groove of the HLA-DR molecule ( Figure 2; Table S4 lists the correspondence between the amino acids at these positions and the six associated classical DRB1 alleles ) . Integrating all of the results , HLA-DRB1*15:01 accounted for 10% of the phenotypic variance in the data , whereas all 6 independent HLA-DRB1 alleles explained 11 . 6% . A model with all identified statistically independent effects ( HLA-DRB1*15:01 , HLA-DRB1*03:01 , HLA-DRB1*13:03 , HLA-DRB1*04:04 , HLA-DRB1*04:01 , HLA-DRB1*14:01 , HLA-A*02:01 , rs9277489/Leu65 in HLA-DPβ1 , rs2516489 , HLA-B*37:01 , and HLA-B*38:01 ) accounted for 14 . 2% of the total variance in MS susceptibility .
We have imputed classical alleles of HLA genes , their corresponding amino acids and SNPs across the MHC , and tested all variants for association in a large case-control collection . Our analysis corroborates the effects of DRB1 alleles other than the well-known DRB1*15:01 association . Classical alleles DRB1*03:01 , *13:03 , *04:04 , *04:01 , and *14:01 display robust , independent associations in our data . The DQB1 and DQA1 genes have been suggested to form extended haplotypes with DRB1 alleles , mostly *15:01 [4] . In our hands , the effect of DQB1*06:02 does not explain the effect of DRB1*15:01 . Furthermore , the DRB1*15:01—DQB1*06:02 haplotype does not appear to explain the data as well as the effect of DRB1*15:01 alone . Based on these results , DRB1*15:01 and the remaining DRB1 alleles are better candidates than DQB1 variants for a causal role in MS susceptibility , a hypothesis that agrees with the MHC analysis of MS subjects with African origin [3] . We note that this interpretation counters evidence in favor of DQB1 from certain murine models that capture elements of human inflammatory demyelination by triggering experimental autoimmune encephalomyelitis induced with myelin-associated oligodendrocytic basic protein [15] or proteolipid protein [16] . A number of studies have highlighted the importance of class I HLA alleles in MS susceptibility , with HLA-A*02:01 being the most prominent allele [17]–[20] . Here , we replicated the HLA-A*02:01 association and attributed it to an amino acid polymorphism at position 95 in the peptide-binding groove of the HLA-A molecule . We also replicated the recently proposed DPB1*03:01 association , and identified a more statistically significant effect at amino acid position 65 in the peptide binding groove of HLA-DPβ1 [7] , [9] . Although our study has overlapping samples with the first study to identify an independent HLA-DPB1 effect [9] , these account for only 24% of the present sample set . The evidence of an HLA-DPB1 effect is strengthened by the fact that the second study reporting such a signal [7] has no overlapping samples with our study . Furthermore , we confirmed the presence of statistically independent HLA-B effects [21] , [22] . Our analysis fine-mapped these to B*37:01 and B*38:01 . Of these , B*37:01 can be explained by amino acid Ser99 of the HLA-B protein , which is also in this molecule's peptide-binding groove . The HLA-C locus demonstrated no convincing evidence for a statistically independent effect , suggesting that previous results may have tagged untested HLA-A or HLA-B effects across the class I region [23] . Although some of the above associations could be explained by specific amino acid polymorphisms in the corresponding HLA proteins , the picture at HLA-DRB1 however appears to be more complex as there was no single model based on amino acids that could explain the entire locus effect ( including the specific effect due to DRB1*15:01 ) . At this stage , our conservative interpretation of these results is that the implicated amino acids allow new hypotheses to be formulated for future functional studies . An interesting finding in our analysis was the association of the region spanning from MICB to LST1 , which contains several important class I , class III and cytokine-related genes . Although the identified SNPs were not significant in the initial ( univariate ) analysis , we established that these reached significance after adjusting for the strong DRB1*15:01 effect . One small study previously examined MICB along with DRB1*15 and had found evidence for an independent association [24] . Another study reported that variation in TNF can modify the effect of DRB1*15:01 [25] . We did not obtain evidence for statistical interaction between this locus and the other MHC variants , indicating that the MHC susceptibility variants we have catalogued likely act independently and additively in terms of MS susceptibility . Overall , we offer robust evidence for the role of a specific MS susceptibility haplotype in this region of the MHC . This region harbors evidence for association with several other diseases , e . g . Crohn's disease and ulcerative colitis [26] , rheumatoid arthritis [27] , Sjogren's syndrome [28] , and hepatitis C virus-associated dilated cardiomyopathy [29] . However , the identity of the causal gene ( s ) within this associated region remains unclear at this time , but it is intriguing that three of the genes ( TNF , LTA and LTB ) are ligands for one of the validated MS susceptibility genes , TNFRSF1A [30] . We did not observe any evidence of statistical interaction ( p>0 . 5 ) with this non-MHC locus in our data . Our preliminary analysis using cis-eQTL data in healthy individuals and MS subjects as well as the publicly available genomic data from the ENCODE and NIH Epigenomics Roadmap did not identify a single variant/gene as the likely causal one . From this information it seems that several genes have functional potential , but more detailed functional studies will be needed to unravel the causal variants and genes . Leveraging genome-wide genotype data , the collection of analyses presented here provides a well-powered investigation of thousands of genotyped and imputed SNPs , classical alleles of 8 class I and II HLA genes and amino acid sequence variation of these HLA proteins . The combination of the large sample size with additional variation types allowed us to present an enhanced dissection of the critical role of the MHC in MS susceptibility . Our results highlight a possible role for certain residues in the peptide-binding groove of HLA molecules associated with peptide antigen recognition . In HLA-DRβ1 we identified a set of four amino acids in positions 71 , 74 , 57 and 86 that capture most ( but not all ) of the DRB1 association . Of these , Val86 has been associated previously with MS [31]–[33] , and this residue appears to be important for the presentation of peptides from a putative target antigen in MS , myelin basic protein [34] , and for the stability of the DRαβ dimer [35] . Another study suggested an association at position 60 [36] and another one at position 13 [37] , although these were not replicated in the present study . Interestingly , the HLA-DRβ1 amino acids in positions 71 and 74 were recently also associated with susceptibility to rheumatoid arthritis [38] . Overall , consistent with the known biology of MS , it appears that disease-associated variants in HLA-DRB1 primarily influence the structural characteristics of the peptide-binding groove and presumably lead to alterations of the T cell repertoire that enhance the likelihood of an inflammatory demyelinating process . However , the MHC also harbors at least one other risk allele that does not directly affect an antigen-presenting molecule: the robust evidence supporting a risk haplotype in the vicinity of MICB will have a different mechanism , one that is likely to affect the function of one or perhaps several cytokines . This study displays an effective strategy for in-depth characterization of this complex region of the human genome . Increasing study sample sizes and more complete reference panels are likely to continue to provide a more detailed perspective on the architecture of genetic susceptibility in this region . The identified amino acid residues may help prioritize the identification of binding peptides and investigations of other potential roles that these susceptibility alleles might have in the biology of MS susceptibility aside from antigen presentation .
We used data from 8 genome-wide association studies ( GWAS ) of European ancestry ( Table 1 ) : ( a ) three GWAS of the GeneMSA [30] , [39] with samples from the Netherlands ( GeneMSA DU ) , Switzerland ( GeneMSA SW ) , and the United States ( GeneMSA US ) ; ( b ) an early GWAS from the IMSGC [30] , [39] , [40] with samples from the United States ( ISMGC US ) and the United Kingdom ( IMSGC UK ) , that was collapsed in one stratum removing the UK cases; ( c ) a GWAS with cases from the Brigham and Women's Hospital and controls from the MIGEN study ( BWH ) [30] , [39]; ( d ) the Australia and New Zealand Multiple Sclerosis Genetics Consortium ( ANZgene ) [41]; ( e ) an unpublished GWAS set from Erasmus Medical Center in Rotterdam , the Netherlands; and ( f ) an unpublished GWAS collection from the Kaiser Permanente MS Research Program ( Kaiser Permanente ) . All the above GWAS data sets were filtered with the same quality control criteria as part of an ongoing meta-analysis of Multiple Sclerosis GWAS . In each of these data sets we performed principal components analysis ( PCA ) to identify population outliers and to calculate covariates to control for population stratification between cases and controls . From each GWAS we extracted SNPs within the extended MHC region ( chr6:29 , 299 , 390 to 33 , 883 , 424; hg18 ) to impute classical alleles for class I HLA genes ( HLA-A , HLA-B , and HLA-C ) and class II HLA genes ( HLA-DPA1 , HLA-DPB1 , HLA-DQA1 , HLA-DQB1 , and HLA-DRB1 ) , their corresponding amino acid sequences and SNPs not captured in the genotypic platforms used . The imputation was performed with the software BEAGLE [42] using a collection of 2 , 767 individuals of the Type 1 Diabetes Genetics Consortium ( T1DGC ) with 4-digit classical allele genotyping for the above HLA genes as the reference panel . This method and reference panel have been used for fine-mapping MHC associations in HIV control [8] and seropositive rheumatoid arthritis [38] . Cases and controls from each GWAS dataset were imputed together . All variants in the reference panel were coded as biallelic markers ( presence vs . absence ) , allowing us to use BEAGLE for the imputation . Post-imputation we excluded variants with minor allele frequency less than 1% from the analysis . Table S5 lists the imputation quality for the identified variants . We analyzed each variant using a logistic regression model , assuming alleles have an additive effect on the log-odds scale . We also assumed the genetic effects were fixed across all eight GWAS . In each model we included the top 5 principal components to control for within-GWAS population stratification and 7 dummy variables to account for between-GWAS specific effects . Throughout the text we refer to such a model as univariate ( Mu ) , even if several covariates were included in the model , reflecting the fact that only one MHC-specific variant is included in the model . This is the representation of the univariate model: ( 1 ) Mu , Univariate logistic regression model where y is the log ( odds ) for the case-control status , β0 is the logistic regression intercept and βi , j is the log-additive effect for the allele j of the variant i with p alleles . In this paper , the term variant is used for any type of SNP ( biallelic , triallelic , etc ) , two-digit HLA allele , four-digit HLA allele and amino acid position . In any case we included p-1 alleles , with the one excluded being the reference allele . Where possible we tried to select the most frequent variant in the controls as the reference allele . The five included principal components are represented in the model as l and the last block in the model represents the dummy variables included for the n studies ( n-1 parameters added in the model ) . To calculate an omnibus p-value for the variant , regardless of the number of alleles included in the univariate model , we used using a log-likelihood ratio test ( 2 ) comparing the likelihood L0 of the null model ( 3 ) against the likelihood L1 of the fitted model: ( 2 ) Log-likelihood ratio test where D is the log-likelihood test value , also known as deviance . D follows an approximate chi-square distribution with k degrees of freedom , where k is the difference of the regressed parameters between the two models . Representation of the null model: ( 3 ) M0 , Null logistic regression model Besides testing variants for association , i . e . SNPs , HLA alleles and amino acids , we also fitted models that estimated the overall effect of the each of the eight HLA genes . We did so , by fitting all respective four-digit alleles of a given HLA gene in the same model . The respective p-values reflect the overall significance of the gene . In order to identify the statistically independent effects , we first tested all variants under a univariate logistic regression model and ranked them based on the p-value of the log-likelihood test . Next , in a forward stepwise fashion , we included in the logistic regression model the most statistically significant variant as a covariate , analyzed all remaining variants and ranked them based on the new p-value of the respective log-likelihood test . The models that included at least one variant as covariate are referred to as conditional throughout the text . In each iteration the null model used in the log-likelihood test was the original null model ( 3 ) with the variants that were used as covariates . We repeated the same steps until no variant or no HLA gene reached the level of statistical significance , which we a priori set to be 10−5 . This statistical significance threshold accounts of 5 , 000 independent tests using Bonferroni correction . Although most of the variants analyzed are correlated , we chose this threshold to account also for the multiple stepwise fitted models . If no variant reached the level of significance but an HLA gene did , we kept adding variants in the overall model until the HLA gene p-value was larger than 10−5 . To compare the effects of two ( or more variants ) , e . g . A and B , we fitted the following models: MA model with variant A , MB model with variant B , and MAB model with both variants A and B . All three model included the same other covariates . Then we used the log-likelihood test to compare MAB vs . MB and MAB vs . MA . These two comparisons represent the effects of variants A and B , respectively , in the presence of the other variant , i . e . B and A . For these comparisons we used the nominal ( α = 0 . 05 ) level of statistical significance . After adjusting for the most statistically significant variant , DRB1*15:01 , the residual effect of the DRB1 locus , i . e . the effect of all alleles besides *15:01 , was still the most statistically significant of any of the remaining variants . This led us to the hypothesis that several other DRB1 alleles could explain the overall DRB1 locus effect , already conditioning on DRB1*15:01 . To identify such effects inside the DRB1 locus , we applied the above forward stepwise logistic regression approach to the four-digit DRB1 alleles . , To test the robustness of the results from the forward stepwise regression , we also applied four other statistical methods for variant selection: i ) lasso , [43] ii ) elastic net , [44] iii ) least angle regression , [45] and iv ) forward Stagewise regression . [46] For the lasso and elastic net we selected the largest value of lambda ( l1 ) after 10-fold cross-validation , such that error was within 1 standard error of the minimum mean cross-validated error . In the respective results section , we illustrate that all methods reached the same conclusion independently . It has been proposed that extended DRB1*15:01–DQB1*06:02 haplotypes confer the risk for MS rather than individual HLA alleles . To test this hypothesis , we used the post-imputation phased data to estimate the DRB1*15:01–DQB1*06:02 diplotypes . Then we fitted a logistic regression that estimated the effect of the diplotype under a per-allelic model . Since this approach used phased data , rather than post-imputation probabilities , the imputation uncertainty is not properly accounted for . Thus , we expect the respective p-values to be slightly inflated . To investigate the functional potential of the MICB-LST1 region we queried: We used Nagelkerke's pseudo-R [47] to estimate the variance explained ( 4 ) Nagelkerke's pseudo-R2 where L0 and L1 are the likelihoods of the null model and fitted model respectively , and N is the number of individuals . We used PLINK for the initial analysis of the data and to estimate minor allele frequencies and imputation quality metrics , i . e . INFO score . [48] We fitted all models in R using the glm function and package lars and glmnet . This investigation has been approved by the Institutional Review Board of Partners Healthcare; the reference number is 2002p000434 . | Multiple sclerosis ( MS ) is an inflammatory and neurodegenerative disease with a heritable component . Although it has been known for a long time that the strongest MS risk factor maps to the major histocompatibility complex ( MHC ) on chromosome 6 , there are still many unresolved questions as to the identity and the nature of the risk variants within the MHC . Because the MHC has a complex structure , systematic investigation across this region has been challenging . In this study , we used state-of-the-art imputation methods coupled to statistical regression to query variants in the human leukocyte antigen ( HLA ) class I and II genes for a role in MS risk . Starting from available SNP genotype data , we replicated the strongest risk factor , the HLA-DRB1*15:01 allele , and were able to identify 11 independent effects in total . Functional studies are now needed to understand their mechanism in MS etiology . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | Fine-Mapping the Genetic Association of the Major Histocompatibility Complex in Multiple Sclerosis: HLA and Non-HLA Effects |
Hyaluronan is a polyanionic , megadalton-scale polysaccharide , which initiates cell signaling by interacting with several receptor proteins including CD44 involved in cell-cell interactions and cell adhesion . Previous studies of the CD44 hyaluronan binding domain have identified multiple widespread residues to be responsible for its recognition capacity . In contrast , the X-ray structural characterization of CD44 has revealed a single binding mode associated with interactions that involve just a fraction of these residues . In this study , we show through atomistic molecular dynamics simulations that hyaluronan can bind CD44 with three topographically different binding modes that in unison define an interaction fingerprint , thus providing a plausible explanation for the disagreement between the earlier studies . Our results confirm that the known crystallographic mode is the strongest of the three binding modes . The other two modes represent metastable configurations that are readily available in the initial stages of the binding , and they are also the most frequently observed modes in our unbiased simulations . We further discuss how CD44 , fostered by the weaker binding modes , diffuses along HA when attached . This 1D diffusion combined with the constrained relative orientation of the diffusing proteins is likely to influence the aggregation kinetics of CD44 . Importantly , CD44 aggregation has been suggested to be a possible mechanism in CD44-mediated signaling .
Hyaluronic acid ( HA ) also known as hyaluronan is a natural carbohydrate polymer constituted by a repeating disaccharide of glucuronic acid ( GlcUA ) and N-acetylglucosamine ( GlcNAc ) ( [-β ( 1 , 4 ) -GlcUA-β ( 1 , 3 ) -GlcNAc-]n ) [1] . Reaching molecular weights of up to 106 Da ( i . e . , several thousand disaccharides ) , HA acts as a space-filling agent , molecular lubricant , and cell migration promoter in processes such as leukocyte trafficking , modulating embryonic morphogenesis and tumor metastasis . It is also an integral component of both the extracellular and pericellular matrices , where it interacts with cells through HA binding proteins [2] , particularly with CD44 [3] . CD44 is a type I transmembrane receptor protein , with HA as its main ligand . It is expressed in a wide variety of human cell types , including leukocytes , endothelial cells , and fibroblasts [3] . Structurally , the canonical form of human CD44 consists of 723 residues , divided into four distinct domains: the extracellular HA binding and stalk domains , the transmembrane domain , and the cytosolic region . From these , only the 158-residue HA binding domain ( HABD ) has been structurally characterized [4] . As its name implies , the majority of the HA-binding capacity of CD44 stems from the globular HABD which can even be expressed as an individual soluble protein that retains its ability to bind HA [5] . It is composed of a link module , which is extended by additional N-terminal and C-terminal flanking regions , that together form a globular HA-binding unit stabilized by three disulfide bridges [4] . The link module itself is a conserved α/β-fold shared by other similar HA binding proteins , such as TSG-6 [6] and LYVE-1 [7] . The CD44-HABD has been studied extensively for two decades [8 , 9] . In addition to a high number of experiments concentrating on its pathophysiology , numerous studies have also focused on the molecular level details of its structure and function [4 , 10–14] . The goal has been to fully understand the factors affecting the CD44-HABD–HA interplay . On the one hand , HABD can recognize HA from other carbohydrates , rendering their interaction highly specific [10] . On the other hand , HABD seems to possess the ability to regulate its affinity to HA , displaying clear differences in affinity between different cell types [15] . These changes in affinity have largely been attributed to N-glycosylations [16 , 17] on the HABD surface , yet additional affinity-modifying mechanisms , such as conformational changes , do also exist [11 , 12] . The binding ability may also be regulated in part by avidity-modifying mechanisms , such as the aggregation of the CD44 receptors [18 , 19] . Similar regulation has been observed with LYVE-1 [20] , an HA receptor homologous to CD44 . Furthermore , the high molecular weight HA has been shown to induce the aggregation of CD44 receptors , while low molecular weight HA seems to lack this ability [21 , 22] . Currently , there is one crystal structure of CD44-HABD in complex with HA [10] . It shows how the carbohydrate ligand binds to a shallow binding groove on the surface of the link module of HABD . This crystallographic binding is characterized by multiple specific hydrogen bonds along with a complementary surface topology of the protein , such as a small hydrophobic pocket able to accommodate the methyl group of a bound GlcNAc residue . The crystallographic study by Banerji et al . also shows two distinct conformations related to the nearby R41 side-chain . The ligand binds HABD with the same crystallographic binding mode in both conformations . In the so-called A-form , the R41 side-chain points outwards from the protein center , while in the B-form it is flipped towards the bound ligand , rendering direct hydrogen bonds possible with it . For this reason , the B-form has been suggested to represent the high-affinity conformation for HA binding [14] , although the exact sequence of events related to the binding process remains unclear . The fact that the crystal structure of HABD without HA displayed only the A-form suggests that the B-form is stabilized by HA binding [4] . In any case , it is widely accepted that R41 is a particularly important residue in the recognition of HA , as mutating it to alanine completely abolishes HA binding [4 , 8 , 10 , 23] . Its crucial role in the recognition has also been confirmed in multiple in silico assays [13 , 14 , 24] . While the study by Banerji et al . shows HA to bind exclusively to the binding groove on the link module [10] , other studies , using both truncation and site-directed mutations , have identified binding residues outside the binding groove to be important for HA binding , too [8 , 23] . For instance , the earliest attempt by Peach et al . [8] to map the HA binding surface of CD44 found multiple arginine and lysine residues located at two clusters—one in the link module ( R29 , K38 , R41 ) and another in the C-terminal extension ( R150 , R154 , K158 , R162 ) —to be crucial for the binding . Especially the residues at the C-terminal extension pose an apparent conflict with the findings of Banerji et al . , as they are structurally distant from the binding groove occupied by HA in the crystal structure . Some of the binding residues mapped to the link module , such as K38 , are also located outside this binding groove , and therefore in conflict with the view proposed by the crystallographic study . In another mutation assay , Bajorath et al . [23] found nine HABD residues to be important for HA binding . First , residues R41 , Y42 , R78 , and Y79 located in the binding groove were found to be vital for HA binding , which agrees well with the crystallographic view . Second , additional residues outside the binding groove ( K38 , K68 , N100 , N101 , and Y105 ) were identified as important for HA binding . Providing further support for these observations , two NMR assays recorded high chemical shift changes upon ligand binding in regions close to these residues [4 , 25] . Overall , mapping all the identified binding residues onto the surface of HABD reveals a widespread interaction surface that cannot be covered by a single rod-like HA polymer . Providing a partial explanation for the above dilemma , previous NMR experiments found a conformation shift in the C-terminal extension of HABD [4 , 11 , 12 , 26] . This shift involves partial unfolding of the C-terminal flanking regions of HABD , thereby excluding the stable link module . In the ordered ( O ) conformation , the C-terminal β9 strand runs anti-parallel to β8 , so that residues after β9 ( 158–169 ) go under the β7–β8 loop [11] . In the partially disordered ( PD ) conformation , the β9 strand unfolds and the orientation of the β8 strand changes with respect to the β0 strand . The existing crystal structures , such as the one resolved by Banerji et al . , assume the O conformation [10] , while several available NMR structures are seen in the PD form [4 , 11] . Ogino et al . demonstrated that the C-terminus interconverts between these two conformations with an exchange rate of hundreds of milliseconds [12] . Furthermore , they also showed that this spontaneous conversion is not dependent on HA binding , although the PD conformation becomes more favorable in the ligand-bound form of the protein . This suggests that while crystallographic structures favor the ordered conformations independently of the presence of HA , NMR measurements in solution capture the PD conformation primarily when HA is present [12] . Recently , Favreau et al . employed the existing 3D structures and computer simulations to provide a structural reasoning for the preference of the PD conformation over the O conformation in the HA-bound case [14] . The conformational freedom gained by the C-terminal residues ( R150 , R154 , K158 , and R162 ) in the PD conformation allows them to attach to the bound ligand that would otherwise lie distant from these residues . This provides additional stabilization of the complex . The O-to-PD transition gives an explanation as to why the C-terminal residues were seen important for HA binding in the earlier mutation assays . However , it does not explain why some residues outside the binding groove and also C-terminal extension have been identified to bind the ligand . The most distinct example of these residues is K38 , identified as important for HA binding in multiple studies based both on alanine-scanning mutations and NMR chemical shifts [4 , 8 , 23] . Providing a complementary explanation to the wide-spread nature of the binding residues able to account for residues such as K38 , Teriete et al . proposed the possibility of multiple different binding modes ( i . e . , CD44-HABD–HA interaction conformations ) covering different regions of the HABD surface [4] . They speculated that in one mode HA could lie in the binding groove , while a second , upright mode could occupy a region perpendicular to the binding groove . It would extend from the C-terminus towards the β4–β5 loop , while passing through the region formed by R41 , Y42 , R78 , and Y79 . Several structures of CD44 have been deposited to Protein Data Bank ( PDB ) [4 , 10 , 11 , 27] , triggering multiple computational simulation studies . These studies have shed light into the molecular level details of CD44 and its interaction with HA . For instance , one study reported a significant immobilization of the monosaccharide units of HA in the binding groove when bound in the crystallographic manner [28] . Water was also found to be severely restricted around the binding residues of HABD , in particular upon complexation with the ligand [29] . Three computational studies also focused on characterizing the A-to-B conformational transition observed in the crystal structures of HA-bound HABD . First , Jamison et al . conducted a comprehensive study elucidating the R41 side-chain dynamics and further characterizing the A-to-B conformational transition initially observed by Banerji et al . [13] . They also discovered that the A-to-B switch in the side-chain of R41 originates from a change in the ϕ backbone dihedral of the adjacent Y42 . Meanwhile , Plazinski et al . elucidated the nature of the interactions in the crystal structure of HA8–CD44 complex . They came to the same conclusion that the A-to-B conformational transition stems from the ϕ backbone dihedral of Y42 [24] . Later , adaptive biasing force sampling was used by Favreau et al . to show that the B-form is energetically more favorable in the ligand-bound receptor compared to the A-form [14] . Two computational surveys also focused closely on the O-to-PD transition [14 , 30] . First , Favreau et al . found the A-to-B and O-to-PD transitions to lack any allostery , and showed how the C-terminal extension gains freedom to bind HA in the PD conformation [14] . Second , Plazinski et al . probed the initial stages of the O-to-PD transition , using the umbrella sampling technique to measure the free energies of the uncoiling of the C-terminal end in wild-type CD44 and Y161A CD44 mutant [30] . This mutant has been observed to adapt exclusively to the PD conformation [12] . Yet , the results of Plazinski et al . showed only minimal differences between the wild-type and the mutant [30] . Finally , Faller et al . glycosylated HABD in silico at two N-glycosylation sites ( N25 and N120 ) to probe the effect of N-glycans on the function of HABD [31] . They concluded that the negatively-charged sialic acids on the termini of the N-glycans charge paired with basic residues important for HA binding , such as R41 or R154 . These sialic acids could thereby impede the binding of HA by competing for the same binding sites [31] . Despite providing valuable insight into the structure and dynamics of CD44-HABD , most of the aforementioned simulations lack the HA ligand entirely or sample only the crystallographic binding mode . Therefore , while providing detailed views of the studied processes , they have not sampled the complete binding process between CD44-HABD and HA . In this work , we characterize , for the first time , all stages of the CD44–HA binding process . To this end , we performed different sets of atomistic molecular dynamics ( MD ) simulations that were designed to probe all the different stages on equal footing . To our surprise , the data revealed the existence of three well-defined binding modes ( Fig 1 ) : Subsequent analysis of the simulation data shows that the crystallographic mode is the strongest of these three modes . However , it is not the most easily accessible and therefore not the most frequently binding mode at the initial stages of binding . The weaker modes foster the aggregation ( in terms of kinetics ) by fostering CD44 diffusion along HA . In this context , our findings provide a solid base to understand the molecular details of the CD44–HA interplay and especially the role of the arginines important in the recognition of HA . In particular , our work shows why and how R41 is crucial in the recognition of HA in all the three binding modes of which one was detected previously by crystallography [10] and two were observed for the first time in our simulations . Our results expand the molecular-scale knowledge of the CD44–HA interplay and therefore have potential to facilitate the development of novel therapies against conditions such as the metastasis of tumors [32 , 33] .
We ran the simulations with the GROMACS 4 . 6 . 7 simulation software package [34] , employing the AMBER99SB-ILDN [35] force field for CD44-HABD and the GLYCAM06h [36] force field for the HA oligomers . These force fields were chosen because they have been shown to capture realistic protein and carbohydrate dynamics and are also compatible with each other [36] . All simulation models used rectangular boxes with periodic boundary conditions . Cubic boxes were used in all the simulations except for the ‘Clustering’ systems , where one dimension of the box was elongated . In these cases , the HA polymer was restrained from its ends ( see S1 File for details ) . All bonds were constrained using the LINCS [37] algorithm , allowing 2 . 0 fs integration time steps . Electrostatics were treated with the particle-mesh Ewald ( PME ) [38] method , with a 1 . 0 nm cut-off distance for the real part . A cut-off of 1 . 0 nm was applied for van der Waals interactions . Furthermore , we applied long-range dispersion corrections for energy and pressure [39] . Neighbor searching was carried out at every 10 steps . All our systems were simulated in the NpT ensemble . The V-rescale [40] thermostat was used to couple the system to a heat bath of 310 K with a time constant of 0 . 1 ps , while the Parrinello-Rahman [41] barostat was employed to couple the system to a pressure bath of 1 . 0 bar , with a time constant of 1 . 0 ps . At the beginning of each simulation replica , we assigned random initial velocities to the particles from a 310 K Boltzmann distribution . We set the saving rate for trajectories to 1/100 ps . All other non-specified parameters used GROMACS 4 . 6 . 7 defaults . To avoid clashes from the construction , before production runs , we minimized the energy of each constructed system with the steepest descent algorithm for 1000 steps and equilibrated the systems for another 2 ps .
Puzzled by the spread of the observed binding residues in the HA–CD44 interaction , we hypothesize that several binding modes may coexist in CD44-HABD . To this end , we performed two sets of simulations ( Table A in S1 File ) : the ‘Seeding’ simulations ( ≈6 μs ) where we placed the HA close to ( 1 nm ) , but not in contact with the binding groove of CD44-HABD; and the ‘Unbound’ simulations ( ≈10 μs ) where we placed the HA far ( 4 nm ) from R41 . Therefore , while HA is in principle biased to find the crystallographic binding site in the ‘Seeding’ simulations , it may explore the surface spontaneously without any bias in the ‘Unbound’ simulations . The end structures of all our simulations were found to correspond to three distinct HA–CD44 binding modes . One of these , referred to as the crystallographic mode , has previously been resolved with crystallography [10] . We also observe the upright mode that has been partially proposed in the literature [4] , along with a new , so far uncharacterised type of CD44–HA complex that we name the parallel mode ( Fig 1 ) . If we orient the CD44 HABD with the N- and C-termini pointing down and R41 pointing towards the viewer , which is the usual orientation in our figures , the crystallographic mode can be recognized from the HA strand as being anchored between the β4–β5 loop ( for naming of the secondary structure elements , see Fig A in S1 File ) and the tip of the side-chain of R41 , located at the α1–β1 motif ( Fig 2 ) . This region is also known as the crystallographic binding groove . The HA strand of the parallel complex , on the other hand , lies on top of R41 and in a lower region of the binding groove , lacking contacts with the β4–β5 loop . Both of these binding complexes share the same orientation of the ligand , i . e . , the reducing end of HA locates on the right side of the figures . In the parallel mode , however , the HA is tilted counter-clockwise to the viewer ( Fig 2 ) , covering a different region of the protein surface . Finally , in the upright complex , HA locates on the right side of R41 and assumes a vertical orientation , with the reducing end of the HA oligomer pointing upwards ( Fig 2 ) . Our findings suggest that HA has considerable interactions with CD44 in three distinct binding modes . This offers a plausible explanation for the wide spatial distribution of the CD44 amino acids critical to HA binding [4 , 23] . Furthermore , visualization of all of our simulations showed only a few transitions between the binding modes , and they mostly occurred soon after the binding had initially taken place . These infrequent transitions may reflect failed docking events between HA and CD44 at early stages of the simulations . Hence , we conclude that the binding modes we observed are stable in the microsecond time scales of the simulations . The free HA ligands in our simulations almost exclusively end up interacting with the crystallographic binding groove or the regions next to it . In the ‘Seeding’ simulations , where the HA16 ligand was placed roughly 1 nm from the protein surface to the crystallographic binding groove , the ligand ended up binding with the protein in three out of four replicas in the parallel mode . In the fourth replica , we observed the formation of the crystallographic complex , which later ( after 200 ns of simulation ) detached spontaneously . However , after 1 μs of simulation , HABD and HA formed the upright complex . Similar binding events were also observed in the five ‘Unbound’ simulations , where the ligand–R41 distance was initially set to 4 nm . In just 400 ns , two replicas formed the parallel binding complex , two formed the upright complex , and one remained less stably bound although still interacting with R41 , highlighting the significant role of this crucial binding residue . Overall , the parallel binding mode was observed to be the most frequent in our unbiased simulations , while the crystallographic mode was seen only once in these simulations . The abundance of the parallel and upright binding modes in the initial stages of the contact suggest that they act as a precursor for stronger binding . It is unclear whether other factors , such as the size , mobility , or orientation of the ligand play any role in determining the binding mode . It is also possible that conformational changes , such as the unfolding of the C-terminal region [12] , or post-translational modifications such as N-glycosylation [42] might affect the preference of the available binding modes . Table 1 lists the previously identified HA-binding residues of CD44-HABD . To allow comparison with the present simulation results , it also shows the percentage of simulation time these residues were in contact with HA in our simulations for each binding mode . The Banerji et al . study found 13 residues of HABD to make prominent contacts with the bound HA: R41 , Y42 , C77 , R78 , Y79 , I88 , N94 , I96 , C97 , A98 , A99 , H101 ( N101 in human sequence ) , and Y105 [10] . These residues are exclusively located in the crystallographic binding groove and form a coherent surface under the bound ligand . Our data from the crystallographic binding mode correlate well with these residues ( see Table 1 ) , excluding N94 and H101 ( N101 in human CD44 ) which are both located at the β4–β5 loop . The reason for these minor differences most likely lies in the slightly different folding of this loop in the murine structure ( PDB:2JCQ ) determined by Banerji et al . [10] and the human protein ( PDB:1UUH ) [4] used in the present study . Earlier , Peach et al . discovered two clusters of binding residues , one in the link module and another in the C-terminal flanking regions [8] . The former comprises residues R29 , K38 , and R41 . While R41 was clearly demonstrated through alanine mutation to be crucial for HA binding , both R29 and K38 also shared a moderate effect on the recognition . Residue K38 lies close to both the binding groove and the C-terminus . In our simulations , this residue realizes contacts with HA exclusively in the upright mode ( 79 . 3% of the aggregate simulation time , see Table 1 ) . On the other hand , R29 maps to the other face of the protein and is not observed to establish contacts with the bound HA in any of our simulations . The second binding cluster that Peach et al . discovered is located in the C-terminal extension of the HABD . It is composed of residues R150 , R154 , K158 , and R162 . The mutation of these residues to alanine results in a moderate decrease in HA binding . However , compound mutations , such as the K158A/R162A double mutant , caused a much more notable decrease . In our simulations of the crystallographic mode , R150 interacts with HA in 88% of the frames . In the parallel mode , R154 binds to HA in 80 . 8% of the frames . Finally , in the upright mode , R162 is one of the primary binding residues interacting with HA in 98 . 2% of the simulation frames . In their crystal structure , Banerji et al . [10] also observed R150 residue to bind to HA transiently but attributed a quite small role for these transient interactions . In light of our simulation data , these flanking residues seem to be important , although each is contributing to a different binding mode . This could explain why Peach et al . found these basic residues to cause a noticeable decrease in HA binding only as compound mutations . Later , Bajorath et al . employed the first in silico model of CD44-HABD to select potential residues for a mutation assay [23] . As a result , nine residues were found to be significant for HA recognition: K38 , R41 , Y42 , K68 , R78 , Y79 , N100 , N101 , and Y105 . The majority of these residues are in line with the Banerji et al . study [10] . For instance , Bajorath et al . found residues R41 , Y42 , R78 , and Y79 at the center of the crystallographic binding groove to be vital for HA binding [23] . Further , supporting their role , these residues are also highly connected to HA in all of the three binding modes studied here , see Table 1 . Additionally , K38 was identified as important for HA binding , agreeing well with the findings of the Peach et al . study . Only the role of K68 was left elusive , as it maps to the other side of the HABD as the binding groove or the ordered C-terminus [10] . It was not found to bind HA in any of our systems either . Given its proximity to R29 that was identified in the Peach et al . study , it is plausible that these residues are linked to some other affinity-modifying mechanism , such as the aggregation of receptors . In addition to the mutation assays , NMR has been used to identify HA binding residues in human CD44 HABD . For example , Takeda et al . identified residues in the binding groove ( T76 , C77 , R78 , Y79 , G80 , I96 , C97 , A98 , A99 ) to be masked by the presence of the ligand , implying that the ligand might have been bound in the crystallographic manner [25] . However , they also noted large chemical shift changes upon HA binding in residues R41 and K38 , from which the latter is not involved in the crystallographic binding . This could imply that other binding modes , such as the upright mode , were present in their HA–HABD constructs . Furthermore , Takeda et al . observed large chemical shift changes in the C-terminal region of HABD ( β8 and β9 sheets ) [11 , 25] . They attributed these changes to the partial disordering of the C-terminus . Teriete et al . observed somewhat a similar rearrangement at the flanking regions of HABD [4] . They also evaluated HA binding residues in HABD based on their chemical shifts . Residues K38 , G40–I44 , R154–Y155 , and N164/E166 gave the most prominent signals , while noticeable changes were also observed with residues R78 , Y79 , R150 , R29 , and R162 , thereby largely agreeing with our observations and with either the Peach et al . or the Bajorath et al . study . The unfolding of the C-terminus from the O to the PD conformation can also explain why the basic C-terminal residues were found to be important in HA binding in the above studies , most notably in the Peach et al . study . As Favreau et al . showed , the C-terminus gains considerable conformational flexibility in the PD conformation , allowing the basic residues to readily come into contact with the bound HA ligand [14] . Hence , this can explain why Ogino et al . observed CD44 population to favor the PD conformation in the ligand bound state [12] . However , regardless of the changes in the C-terminus , several of the above studies have listed K38 as an important binding residue even though it is not located in the binding groove or in the unfolding C-terminal extension . This finding combined with our observation that K38 is important only in the upright mode could imply that there exists additional HA–CD44 binding modes outside the crystallographic one . There are five N-glycosylation sites in CD44-HABD ( N25 , N57 , N100 , N110 , and N120 ) [4] , so the presence or absence of N-glycans might also alter the binding . Furthermore , English et al . have shown that all the five glycosylation sites on HABD host N-glycans when expressed with cancer cells [16] . Several studies have probed the effect of mutating these N-glycosylation sites on HA binding [15–17 , 43] . For instance , mutating N25 or N120 to serine increased HA binding considerably , whereas mutating the other three sites displayed only a negligible effect [16] . On the other hand , Bartolazzi et al . showed that mutation of any of the five N-glycosylation sites is enough to abolish HA binding [43] . The contradictory shreds of evidence might be explained by , for example , cell-type specific differences . However , due to these discrepancies , the role of the N-glycosylation site residues for HA binding remains unclear . In our simulations , the non-glycosylated N25 and N110 were both observed to be in contact with HA over a half of the simulation time in every binding mode , while the rest of the N-glycosylation sites made fewer contacts with HA . Considering that Peach et al . and Bajorath et al . found residues ( R29 and K68 ) close to N120 and N57 , from which N120 was deemed important in the English et al . study , it is possible that the existence of N-glycans gives rise to or amplifies some binding modes . It is also worthwhile to notice that the studies listed in Table 1 use different expression platforms for their CD44-HABD constructs , which might differ considerably in their glycosylation content , thereby affecting the results obtained . All in all , only five residues in Table 1 ( R29 , N57 , K68 , G80 , N120 ) lack any HA-interaction over 30% in our simulations . These residues map to the “back” side of HABD . All the other residues , locating at the “front” face ( i . e . , the face of the binding groove ) of HABD have over 30% contact time with the ligand in at least one binding mode . Hence , the existence of multiple binding modes seems a plausible explanation for the observed spread of the HA-binding residues . Our simulations showed that three different binding modes establish spontaneously between HA and CD44 . To further characterize the structural and dynamical features of these binding modes , we performed the ‘Gathering’ simulations , where each simulation was prepared in a well-defined binding mode: crystallographic , parallel , or upright ( Fig 2 ) . In total , three independent replicas for each binding mode were simulated for 1 μs . The binding mode remained unchanged in every one of them . For detailed descriptions of the structural features of each binding mode , see section 3 in S1 File . In Fig 3 , we depict for each binding mode the contacts between HA and every arginine residue of CD44 . They display a unique contact signature or fingerprint , which correlates well with the regions of lesser HA mobility in Fig 4 . Hyaluronan-binding proteins usually contain arginine-rich motifs [2] and their mutations often substantially influence the hyaluronan binding affinity [8] . In the case of CD44 , these arginines are very widespread on the surface of the protein . One possible explanation for this counter-intuitive observation is the existence of several binding modes . Despite the observed clear differences between the binding modes ( see also Figs D to O in S1 File ) , they also share common features . To begin with , all binding modes share a strong interaction with R41 and the neighboring Y42 . Without exceptions the contact region around R41 and Y42 is the least mobile of all the HA–HABD contacts , denoting clear binding stabilization , see Fig 4 . This stabilization illustrates why R41 mutations lead to a considerable decrease in the binding affinity despite the observation that the CD44-HABD 3D structure seems unaffected in complementary antibody assays [8]; any change around R41 would perturb all three binding modes . Moreover , the fact that this modification strongly inactivates HA binding suggests that the three binding modes found in this work may be the only significant ones existing [8] . Another important feature shared by all the binding modes is the extended hydrogen bond network region towards the reducing end of the bound HA , between T108 and Y114 . Finally , the interaction with R78 is also shared by all three modes , while it seems to be particularly prominent only in the crystallographic and upright modes . While all binding modes share features that can be considered to be the core of their binding in the CD44-HA complexes , our results show that each mode also has other interactions that partially stabilize their structure . In each binding mode , our data show that HA establishes hydrogen bonds with at least another basic residue placed between 3–4 nm towards the non-reducing end ( see Fig I in S1 File ) : R150 for crystallographic; R154 and partially R150 for the parallel; R162 for the upright mode . This common feature seems to play an important role in stabilizing the mobile end of the non-reducing HA end , as can be seen in Fig 4 . These two flanking arginines R41-RX ( where X depends on the binding mode ) anchor the HA oligomer to the surface of CD44-HABD with 6 to 8 monosaccharides , thereby increasing the stability of binding . This provides a reasonable explanation to why the binding is severely diminished with very short fragments of HA , e . g . , HA4 [4] . There is conflicting experimental evidence regarding the stabilization by the flanking arginines . While Peach et al . found decreased binding upon mutation ( R150A , R154A , R162A ) consistent with our results [8] , Banerji et al . found that the same mutations have no effect on HA binding in a HA-ligand binding assay [10] . The source of this contradiction is currently unknown but our data align with the observations of the former study . It is possible that the parallel and upright binding modes represent metastable states , as described in Ref . [44] , preceding the crystallographic complex . Supporting this , the upright and especially the parallel mode are relatively abundant in the timescales of our simulations , while the crystallographic form is seen to form only twice . On the other hand , we did not observe any direct transitions between any of the binding modes . Given our approximately 50 microseconds of sampling , this implies that such transitions are rare in the timescales of simulations . Taking into consideration that the time scales probed here are short compared to the biologically relevant ones , we proceeded to study the free energy profile of each binding mode to quantitatively assess their importance . We computed the free energy profiles of detaching the HA from the HABD surface with the umbrella sampling technique . We investigated the binding free energy of HA8–HABD in two cases: the parallel complex and the crystallographic complex , see Fig 5 . For the parallel form , we obtained a free energy difference of −22 kJ mol−1 ( ∼8 . 8 kBT ) , while the crystallographic mode indicated a clearly stronger value of −33 kJ mol−1 ( ∼13 . 2 kBT ) . The orders of magnitude of these values are in agreement with experimental observations , where attachment of HA oligomers of this size to CD44-HABD has been found to be reversible [18] . Our estimates for the binding strength also correlate well with the experimentally observed values [10 , 18] . The free energy difference between the two binding modes ( ∼4 . 4 kBT ) suggests that the crystallographic mode is 80 times more favorable than the parallel mode , although the latter is more probable in our simulations . If we use the ratio of the integrated populations as obtained from free energy profiles to evaluate the probability [45] of each mode , the difference reduces to ∼20-fold . This implies that the entropic component of the free energy favors the parallel mode over the crystallographic . In either case , the difference of free energies is small enough for both modes to coexist to a substantial degree . The parallel mode might become a plausible alternative especially in a scenario , where the availability of the crystallographic binding site is hindered , e . g . , due to N-glycosylation . Calculating the free energy with the umbrella method has its caveats , see S1 File . Being aware of such limitations , we decided to support our free energy calculations with an alternative measurement method . Inspired by the electrophoresis experiments used to determine the strength of interaction between HA and CD44-HABD [17] , we designed an in silico counterpart . Taking advantage of the total negative charge of HA , we applied an external electric field to the HA–CD44-HABD complex . In our simulation experiments , the CD44-HABD was fixed and oriented so that the HA binding groove was facing the in silico equivalent of the positive electrode ( anode ) . The primary advantage of these simulations is the more natural detachment of the ligand compared to the umbrella free energy simulations , thereby mimicking this biologically relevant process in a more appropriate setting . In practice , we applied two fields with different strengths . In the first set , ‘E-field strong’ , the field was strong enough to ensure that detachments took place in short simulation times ( up to 20 ns ) , see Table 2 . We then performed a total of 20 simulations for each binding mode and calculated the number of contacts between the HA ligand and CD44 . When the number of contacts between HA and HABD reached zero for the first time , we considered the ligand as “detached” . The results ( Table 2 ) obtained agree with our previous calculations . The crystallographic mode seems to be the strongest again , while the parallel mode seems to be the weakest . The upright mode shows intermediate behavior . Interestingly , our data also show that the crystallographic A-form is weaker than its B-form counterpart , displaying a strength similar to the upright mode . This result supports the view that the A-form is just an intermediate state to the B-form , which is further stabilized by hydrogen bonds with R41 [14] . It is important to stress that the strong field applied in this set is certainly unphysical and results in substantial water orientation , minimizing its dipole moment , see Fig C in S1 File . Naturally , this changes the solvation environment of the protein and HA , significantly affecting the binding affinity we aim to measure . Due to the field-induced effects , such as the water orientation , we performed a second simulation set , ‘E-field weak’ , where we reduced the strength of the field to a tenth of its original value . In this set , the orientation of water was significantly closer to the orientation of bulk water in zero-field ( see Fig C in S1 File ) , and therefore the environment of the HA–CD44-HABD complex is here appropriate compared to biological conditions . Due to the decrease in the strength of the field , each simulation was extended to 500 ns , totaling 10 μs of simulation per binding mode . The results with the weaker field ( Table 2 ) are consistent with the other two methods , indicating that our estimates for the binding strength are reliable and robust . Additionally , our ‘E-field weak’ simulations provide interesting information about the intrinsic stiffness of each binding mode . As seen in Fig B in S1 File , where we plot the distance of HA ( the methyl group of GlcNAc ( 0 ) ) to R78 during the detachment simulations , each binding mode displays a unique signature . While the effect of the electric field is not noticeable in the crystallographic complex , the upright complex fluctuates heavily around its initial equilibrium position . We also observed a few events were there is partial detachment . In these cases , the HA strand , however , remained bound through interactions with the residues 108–114 of the protein . This points to the existence of regions which can stabilize or initiate a given binding mode . Finally , we observed that in the parallel mode there is a clear rearrangement or detachment when one applies the field . It is important to mention that in this mode the R78 interaction was not very relevant even though it was used as a reference point in the protein when calculating the distances . We conclude that the sampled binding modes have different free energy profiles . The crystallographic mode is the most favorable one , while the parallel mode is the least favorable . However , we also see that the weaker binding modes , i . e . , parallel and upright , have significantly larger regions of interaction with CD44-HABD , thereby increasing the probability of their occurrence . This is largely the reason why we observed a considerable variation in the free energy when calculated by the difference between the bulk and the minimum in the free energy profile or by integrating the profile independently for the bound and unbound states . Therefore , our results suggest that while the crystallographic mode plays a crucial role in the HA–CD44-HABD interaction , the metastable modes also increase the HA–CD44-HABD binding constant ( sum of the binding constants of each binding mode ) . Furthermore , in real biological systems , one rarely maintains pure equilibrium conditions . The orientation of the molecules , blood flow , or interaction with other molecules might induce an external bias to CD44–HA complexes resting on the plasma membrane and biasing the preferred binding mode . The crystallographic binding complex can be found in two conformational states , the so-called A- and B-forms , see Fig P in S1 File . While the HA ligand may be bound to both conformations , the A-form lacks direct R41–HA contacts , whereas the B-form enables the R41 side-chain to form two direct hydrogen bonds with the bound ligand [10] . Another computational survey recently discovered the molecular basis for the conformations: the ϕ dihedral in Y42 backbone acts as a bistable switch , altering the shape of the β1–α1 loop region , which in turn causes the R41 side-chain to flip between these two conformations [13] . Furthermore , many studies suggest that the presence of HA in the binding groove stabilizes the B-form [10 , 13 , 14] . We found the spontaneous formation of the crystallographic B-form complex in two occasions , see supplementary video . Importantly , these events occurred in equilibrium simulations , meaning that they were not initiated by external perturbations . Instead of binding directly to the crystallographic B-form structure , we found on both occasions that HA binds first to the A-form structure . The B-form complex formed only when HA had aligned correctly to the binding groove , a process most distinctively characterized by the methyl group of GlcNAc ( 0 ) pointing to the hydrophobic pocket . Given these data , the formation of the crystallographic complex is described by the following equation: This finding is in line with the current predictions and implies that the crystallographic A-form complex is an intermediate state in the formation of the B-form complex . Providing further support for this notion , we did not observe spontaneous B-to-A flips in the HA-bound systems once the B-form had established . We have , however , recorded such transitions in ligand-free simulations or in cases where the ligand has detached ( see Fig Q in S1 File ) . This observation is in line with previous computational studies that revealed the type-B conformation to be the favorable conformation in the ligand bound HABD [14] . In terms of the interactions with the ligand , the differences between the A and B complexes are subtle . In fact , the lack of contacts with the R41 side-chain is the only notable difference between the two conformations ( Fig 3 ) . As already shown by Favreau et al . in a previous computational study , the flipping of the R41 side-chain suffices to provide additional stabilization for the complex . This is supported by the data in Table 2 , which show that the A-form complex dissociates more often under the electric field than the B-form complex . Finally , it is important to stress that the A and B conformations are only relevant with the crystallographic binding mode . In the parallel mode , for instance , the transition cannot happen as HA lies on top of R41 , restricting its movement . HA exists in a variety of sizes . Its biological effects depend on its molecular weight [46] . Low molecular weight HA ( i . e . , up to 20 carbohydrate units ) has , for example , been implicated in the stimulation of cell proliferation [47] , while high molecular weight HA inhibits the same process [48] . Interestingly , the length of HA has been reported to influence also the aggregation of the CD44 receptors . In a recent study , polymeric HA stimulated the aggregation of CD44 in vivo in multiple cell types [21] . Oligomeric HA comprised of 6–20 carbohydrate units , on the other hand , disrupted this aggregation effect . In another study , increasing the length of HA was observed to augment the CD44–HA binding in vitro , with a saturation point at molecular weights of 262 kDa , corresponding to polymers of almost 700 disaccharide units [18] . Provided that the HA polymers are long enough , incrementing the surface density of CD44-HABDs seems to result in the same effect [18] . Finally , the aggregation affinity depends on the CD44 variant , suggesting that differences in the HA–CD44 interaction also play a role [19] . In light of the above , we elucidated the molecular details of the interaction between a high weight HA polymer and multiple CD44s . In practice , we conducted two replica simulations termed “Clustering” , where we simulated two CD44-HABD proteins together with HA64 , spanning the length of the simulation box ( 30 nm × 8 nm × 8 nm , see Methods in section 2 . 3 in S1 File for details ) . Both HA and CD44-HABD are restrained , however they able to move freely along the long edge of the simulation box ( i . e . , in the direction parallel to the HA polymer ) . Overall , the conditions simulated in this system mimic those present at a cell membrane , with the CD44 receptors protruding from the plasma membrane and interacting with a rod-like HA polymer , like in the pericellular matrix of a cell [46 , 49] . CD44 proteins can move along the HA64 strand . While the first protein ( P1 ) in the first replica simulation remained stationary relative to HA , the second protein ( P2 ) moved a stretch of roughly 20 carbohydrate units along the HA64 strand during the 3 μs simulation ( see the upper graph in Fig 6 ) . This observation illustrates that HA can effectively capture CD44 HABDs resulting in an increasing local CD44 concentration along HA that can foster aggregation . Additionally , HA limits CD44 diffusion into one dimension ( along the HA polymer ) . This dimensional restriction influences the effective contact times and relative orientation of the CD44–CD44 pairs , which can further enhance the aggregation kinetics of these receptors [50] . The diffusion of the protein is possible due to periodic , yet transient ( 10–20 ns ) , partial detachments from HA . In this replica , neither of the proteins remained in any specific binding mode but kept sampling both the parallel and upright modes , the latter being possible due to the tilting of the protein . The most well-defined binding process occurred with the diffusing protein ( P2 ) at 1 . 1 μs ( see Fig 6 ( top ) ) , when the upright complex was formed . However , it lasted only about 400 ns before another detachment . Furthermore , in this simulation , both proteins always interacted spontaneously with HA64 with their R41-containing face , again highlighting its importance in recognition . In the second replica simulation , the protein P3 formed a crystallographic binding complex ( see S1 Video ) . After the initial contact to residues 54–62 of the HA64 polymer , the protein moved a stretch of three carbohydrate units before binding HA in the crystallographic mode . At this stage , the protein was in the type A conformation , but after roughly 700 ns it turned into the type B form , thereby completing the binding . The formation of the A-form crystallographic complex and the commencing of the type B conformation can both be observed as increases in the contacts in the lower graph of Fig 6 . These occur at 750 and 1500 ns in the trajectory , respectively . Importantly , these binding events completely halted the diffusion of the protein relative to HA64 , highlighting the strength of the crystallographic binding complex . The protein P4 , on the other hand , was loosely bound throughout the simulation , sampling both parallel and upright binding modes , analogously to the second protein in the first replica . The above observations provide a plausible explanation for the existence of multiple binding modes with varying affinities . With only the strong crystallographic B-form , the relative movement along the HA polymer would not be possible , and therefore the aggregation of the CD44 receptors ( in terms of kinetics ) along HA might not be enhanced . Supported by the fact that the parallel and upright modes are present in microsecond time scales , our simulation data support the view that a non-zero fraction of the CD44 population is bound to HAs through these metastable binding modes , thereby fostering their diffusion along HA .
Based on an extensive analysis of microseconds of atomistic MD simulation data generated in this study , there are three different binding modes for the CD44–HA interaction . We call them here the ‘crystallographic’ , ‘parallel’ , and ‘upright’ modes . From these mutually exclusive binding orientations , the crystallographic mode is well characterized in the existing literature , while the latter two were observed for the first time in the present study . The fact that we observed these binding modes in this work stems from the system set-up , which allows the components ( i . e . , CD44 and HA ) to move freely in solution . The spontaneous formation of these three binding modes in our unbiased simulations gives an explanation as to why the previous mutagenesis and NMR shift studies [4 , 8 , 23] identified so many topographically widespread CD44 residues to be involved in the recognition of HA . The other possible explanation given in the literature is the partial disordering of the C-terminal extension of HABD . However , there is no obvious reason why these two explanations would rule each other out . Our estimates for the relative binding affinity further revealed that the crystallographic mode , first described in Banerji et al . in 2007 , is the strongest of the three modes . Meanwhile , the parallel mode is the weakest of the three but also the most frequently found orientation in our simulations . However , the differences in binding affinities are quite small and within a range of up to 5 kBT , implying that all the modes can coexist at the same time with non-zero proportions , thus increasing the binding constant . Especially the existence of N-glycans on HABD might readily alter the relative propensity of these binding modes . Based on our work , every arginine residue on the same face of the protein as R41 seems to be useful in stabilizing some of the characterized binding modes . However , R41 is the only residue interacting with every mode , highlighting its importance in the recognition . R78 also participates in all of the modes , however , its contribution to the parallel mode is minimal . Furthermore , R150 , R154 , and R162 are all important for at least one of the binding modes . This contrasts with the results of Banerji et al . who only considered a static view based on the crystallographic binding mode found by X-ray . However , the importance of these flanking arginines might explain why short ( 1–2 disaccharide units ) HA fragments do not bind strongly to CD44 . Finally , the fact that R41A mutation abolishes HA binding suggests that the three R41-dependent binding modes found in this work are the most significant ones . We also further clarified the structural details of the well-characterized crystallographic binding mode . Namely , this is the first study where the spontaneous formation of this binding complex has been recorded , confirming that the HA oligomer first binds to CD44 in the A-form ( ‘open’ conformation ) in a crystallographic manner , after which the B-form ( ‘closed’ conformation ) commences . Upon the process of attachment , CD44 was also observed to diffuse along the HA strand . This suggests that HA can restrict the diffusion of CD44 proteins to one dimension ( to take place along the HA polymer ) which together with the larger local CD44 concentration along HA can potentially promote aggregation ( kinetics ) of these membrane-bound receptors . Our data also suggest that the crystallographic binding complex alone is too strong for the diffusion to take place , thereby providing an additional reason for the existence of the weaker modes . Overall , aggregation is a viable regulation mechanism for membrane proteins . It can , for example , exclude proteins that are aggregated from transmitting signals across the membrane [3] . Although many details of CD44-mediate signaling are still unclear , it could be regulated similarly . Lastly , there is reason to keep in mind that CD44 is ( in its natural form ) a glycoprotein housing five possible N-glycosylation sites in the HABD . It is possible or even likely that the presence of N-glycans can switch the population of proteins to favor some HA binding mode over the others , or even generate totally novel interaction ways for these complex macromolecules . Hence , it is also possible that the lack of N-glycans leads HA and CD44 to interact in ways that are less relevant in mammalian cells . | Hyaluronan is a natural sugar polymer in our bodies . Besides acting as a space-filling agent for example in multiple connective tissues , it also functions as a cellular cue in cancer and inflammation . Our tissues sense hyaluronan through receptors—proteins that sit at the surface of cells and grab the molecules they are expected to recognize . Although the knowledge associated with hyaluronan and its receptors is constantly accumulating , the molecular-level insight is largely missing or incomplete due to the lack of techniques able to probe the dynamics of protein–carbohydrate interactions with sufficiently high resolution . In this work , we characterize the binding of hyaluronan to its receptor CD44 with atomistic precision . We achieve this level of precision by employing atomistic molecular dynamics simulations . This computational technique allows one to follow the movement of atoms of a virtual system at scales beyond the resolution of any experimental technique . Our work specifically focuses on the different stages of hyaluronan–CD44 binding , and we observe the process to involve three different binding modes , making it more versatile than previously thought . Our insights , therefore , promote the understanding of the interplay between hyaluronan and HA , thereby fostering development of new drugs or inhibitors to malignancies , such as cancer metastasis . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"crystal",
"structure",
"chemical",
"compounds",
"cd",
"coreceptors",
"condensed",
"matter",
"physics",
"carbohydrates",
"organic",
"compounds",
"materials",
"science",
"basic",
"amino",
"acids",
"amino",
"acids",
"crystallography",
"coreceptors",
"oligomers",
"macromolecules",
"thermodynamics",
"materials",
"by",
"structure",
"polymers",
"polymer",
"chemistry",
"solid",
"state",
"physics",
"proteins",
"chemistry",
"free",
"energy",
"physics",
"biochemistry",
"biochemical",
"simulations",
"signal",
"transduction",
"arginine",
"cell",
"biology",
"organic",
"chemistry",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"computational",
"biology"
] | 2017 | Atomistic fingerprint of hyaluronan–CD44 binding |
Cancer-associated fibroblasts ( CAFs ) , the most common constituent of the tumor stoma , are known to promote tumor initiation , progression and metastasis . However , the mechanism of how cancer cells transform normal fibroblasts ( NFs ) into CAFs is largely unknown . In this study , we determined the contribution of miRNAs in the transformation of NFs into CAFs . We found that miR-1 and miR-206 were down-regulated , whereas miR-31 was up-regulated in lung CAFs when compared with matched NFs . Importantly , modifying the expression of these three deregulated miRNAs induced a functional conversion of NFs into CAFs and vice versa . When the miRNA-reprogrammed NFs and CAFs were co-cultured with lung cancer cells ( LCCs ) , a similar pattern of cytokine expression profiling were observed between two groups . Using a combination of cytokine expression profiling and miRNAs algorithms , we identified VEGFA/CCL2 and FOXO3a as direct targets of miR-1 , miR-206 and miR-31 , respectively . Importantly , systemic delivery of anti-VEGFA/CCL2 or pre-miR-1 , pre-miR-206 and anti-miR-31 significantly inhibited tumor angiogenesis , TAMs accumulation , tumor growth and lung metastasis . Our results show that miRNAs-mediated FOXO3a/VEGF/CCL2 signaling plays a prominent role in LCCs-mediated NFs into CAFs , which may have clinical implications for providing novel biomarker ( s ) and potential therapeutic target ( s ) of lung cancer in the future .
Fibroblasts , a key cellular component of human tissues and tumors , can be divided into resting and activated fibroblasts [1] . For example , fibroblasts are highly activated at the site of healing wound [2] . The activated fibroblasts invade lesions and generate extracellular matrix ( ECM ) to serve as a scaffold for other cells . Once a wound is repaired , the activated fibroblasts revert to a resting phenotype , which are mainly known as normal fibroblasts ( NFs ) [2 , 3] . It is widely accepted that the development of tumors is not just determined by malignant cancer cells , but also by the activated tumor fibroblasts or carcinoma-associated fibroblasts ( CAFs ) [4] . The CAFs stimulate cancer cell proliferation and progression through the secretion of a variety of cytokines , chemokines and ECM [1 , 5] . Growth factors , for example , vascular endothelial growth factor ( VEGF ) [6] , transforming growth factor-β ( TGF-β ) [7] , and fibroblast growth factor 2 ( FGF2 ) [8] , are believed to play crucial roles in fibroblasts activation . However , the molecular mechanisms of the conversion of NFs into CAFs are poorly understood . MicroRNAs ( miRNAs ) represent a class of small non-coding RNAs with an important regulatory role in various physiological and pathological processes [9] . Accumulating evidence suggests that miRNAs play a regulatory role not only in cancer cells during carcinogenesis but also in the transition or activation of fibroblasts [10 , 11] . For example , deregulation of miR-31 , miR-214 and miR-155 reprogrammed NFs into CAFs in ovarian cancer [11] . Down-regulation of miR-148a in endometrial cancer CAFs stimulates the motility of endometrial cancer cells [10] . Up-regulation of miR-106b in CAFs promotes gastric cancer cell migration and invasion [12] . MiR-21 , a well-known oncomiRNA , was found significantly up-regulated in colorectal CAFs and the latter contributed to colorectal cancer growth and invasion [13] . However , how miRNAs are involved in the conversion of quiescent resident fibroblasts to CAFs in lung cancer remains largely obscure . In our current study , we demonstrated that i ) deregulation of miR-1 , miR-206 and miR-31 contributes to the conversion of NFs to CAFs in lung cancer; ii ) combination of miR-1 , miR-206 and miR-31 reprograms NFs to CAFs through mediating FOXO3a/VEGFA/CCL2 signaling; and iii ) modifying tumor microenvironment via targeting three miRNAs or CCL2/VEGFA significantly reduced tumor angiogenesis , TAMs accumulation , tumor growth and lung metastasis . Our results have clinical implications by providing novel biomarkers for lung cancer diagnosis and may possess therapeutic application for lung cancer treatment in the future .
To compare the miRNAs expression profile in primary human CAFs and NFs , we performed a miRNA assay to profile the global expression of mature miRNAs in 3 pairs of CAFs isolated from lung carcinoma and matched healthy NFs extracted from a normal area of tissue , at least 10 cm from the tumor area . Both CAFs and NFs were fibronectin and vimentin positive cell populations [5] . The expression levels of alpha smooth muscle actin ( α-SMA ) were significantly higher in CAFs compared to NFs ( S1 Fig ) . We found that miR-1 , miR-206 , and miR-31 were among the most significantly down- and up- regulated miRNAs in CAFs compared with those in NFs , respectively ( Fig 1A ) . The down-regulation of miR-1 and miR-206 and up-regulation of miR-31 were further confirmed in 15 paired CAFs and NFs from different lung cancer patients by Taqman qRT-PCR ( S2 Fig ) . More interestingly , consistent with our observations in cancer tissues , we found that circulating miR-1 , miR-206 and miR-31 were also dramatically down- and up- regulated in lung cancer plasma compared with healthy plasma ( Fig 1B ) . In the next step , we compared the expression of miR-1 , miR-206 and miR-31 in NFs , which were isolated from the patient sample ID 1 , co-cultured with or without RFP-expressing A549 or H460 cells . After 10 days of coculture , NFs were isolated by a flow cytometry sorter within negative RFP signaling cell population . The miR-1 , miR-206 down-regulation and miR-31 up-regulation were observed in co-cultured NFs compared to mono-cultured NFs ( Fig 1C ) . Similar alternative pattern of miR-1 , miR-206 and miR-31 can also be observed in NFs from the patient sample ID 7 , 11 , 15 in co-culture ( S3 Fig ) . Furthermore , we discovered that the lung cancer cells ( LCCs ) -reprogrammed NFs exhibited stronger ability to promote lung cancer cell migration and colony formation ( Fig 1D and 1E ) , suggesting that cancer cells impart CAF-like properties to NFs during co-culture . To study the function roles of miR-1 , miR-206 and miR-31 in NFs-CAFs conversion , we triple transfected anti-miR-1 , anti-miR-206 and pre-miR-31 in NFs ( hereinafter referred to as "NFs-TM" ) to modify NFs into CAFs-like fibroblasts . We observed that NFs-TM significantly enhanced migration and colony formation ability of co-cultured LCCs . Similarly , restoration of miR-1 , miR-206 and knockdown of miR-31 levels in CAFs ( hereinafter referred to as "CAFs-TM" ) impaired the ability of CAF-induced migration and colony formation of co-cultured LCCs ( Fig 2A and 2B ) . These results suggested that deregulation of these three miRNAs could promote NFs converting to CAFs or at least to CAFs-like fibroblasts . Similar cell migration and colony formation results were also obtained when CAFs and NFs from sample ID 7 , 11 , 15 were applied in co-culture ( S4 Fig ) . Thus , CAFs and NFs from sample ID 1 will be mainly used in the following experiments . To study the effect of identified miRNAs on NFs-CAFs conversion in vivo , A549 cells were subcutaneously injected alone or co-injected with NFs-Scr , NFs-TM , CAFs-TM , or CAFs-Scr into immunodeficient mice . We found that both NFs-TM and CAFs-Scr dramatically enhanced tumor growth and angiogenesis when compared to NFs-Scr and A549-alone . The CAFs-promoted tumor growth and angiogenesis effect was abolished by up- and down- expression of miR-1 , miR-206 and miR-31 in CAFs ( CAFs-TM ) . No statistical significance in tumor weights was observed between NFs-Scr and A549-alone ( Fig 2C and 2D ) . We assessed the cancer cell and fibroblasts fractions in mouse xenograft tumor by immunofluorescence staining using an antibody specific for human vimentin , which A549 cells fail to express [5] . We found that green fluorescence signal , as an indicator of the fibroblast population , was significantly higher in CAFs-Scr and NFs-TM co-injection than A549-alone , NFs-Scr and CAFs-TM co-injection ( S5 Fig ) . These results suggested that the original fibroblast populations CAFs-Scr , NFs-Scr , CAFs-TM and NFs-TM commingled with cancer cells contributed to the tumors . All fibroblasts survived and even proliferated in tumors together with cancer cells . CAFs-Scr and NFs-TM were more competent in enhancing A549 tumor growth . Similarly , several secreted factors , such as vascular endothelial growth factor ( VEGF ) [6] , stromal cell-derived factor-1 ( SDF-1 ) [5] , Chemokine ( C-C motif ) ligand 5 ( CCL5 ) [14] , Chemokine ( C-C motif ) ligand 2 ( CCL2 ) [15] and matrix metalloproteinase 9 ( MMP9 ) [1] , have been implicated as possible regulators for enhancing tumor growth . Using real-time PCR , we observed increased levels of VEGF , CCL2 and MMP9 in both CAFs-Scr- and NFs-TM co-injection than NFs-Scr and A549-alone ( S6 Fig ) . Furthermore , mice bearing A549+CAFs-Scr and A549+NFs-TM tumors displayed a marked increase in the number of micro- and macro- scopic lung metastases compared with those in A549+NFs-Scr and A549-alone . However , no statistical significance in lung metastasis was observed between A549+CAFs-Scr and A549+CAFs-TM ( Fig 2E ) . Tumor-associated macrophages ( TAMs ) are known to promote tumor progression and malignancy [16 , 17] . To compare TAMs infiltration in different tumor groups , we analyzed the presence of TAMs in single-cell suspensions from tumor tissues by flow cytometry assay for CSF-1R and F4/80 staining , two well-characterized markers of TAMs [18] . We found that TAM infiltration was significantly higher in CAFs-Scr- and NFs-TM co-injection than NFs-Scr and A549-alone , suggesting that interaction of CAFs-Scr and NFs-TM with LCCs facilitates the recruitment of TAMs to lung tumors ( Fig 2F ) . Secreted factors , such as VEGF , TGF-β , HGF and SDF-1 , have been implicated as being important cell co-culture regulators [4 , 19 , 20] . To better understand the crosstalk between NFs , CAFs and LCCs and to determine whether miRNA-reprogrammed NFs-TM can mimic CAFs in co-culture , the conditioned media ( CM ) from NFs-Scr-A549 , CAFs-Scr-A549 and NFs-TM-A549 were screened for various cytokine , chemokine and growth factor levels using the Luminex-based BioPlex suspension array system [14] . As shown in Fig 3A , 6 genes ( CCL2 , CCL5 , IL-6 , IL-8 , bFGF , and VEGFA ) were significantly up-regulated ( at least 2 . 5-fold ) in CM from CAFs-Scr-A549 and NFs-TM-A549 as compared with CM from NFs-Scr-A549 . We noticed that the gene expression profile of NFs-TM-A549 was more similar to CAFs-Scr-A549 than NFs-Scr-A549 , indicating miRNA-reprogrammed NFs-TM possesses some properties of CAFs . The top 5 soluble factors that demonstrated the highest up-regulation from cytokine expression profiles were selected for further study ( CCL2 , CCL5 , IL-6 , IL-8 , and VEGFA ) . To assess the contribution of each or a combination of the soluble factors to the functional role of lung cancer cells ( migration and colony formation ) , LCCs were treated with a single or a combination of secreted factors . We observed that both single and combination treatment of secreted factors ( CCL2 , CCL5 , IL-6 , and IL-8 ) predominantly affected cancer cell migration but had no effect on colony formation ( Fig 3B and 3C ) . The LCCs migration and colony formation were dramatically enhanced by the combination of ( CCL2 and CCL5 ) , but not ( IL-6 and IL-8 ) , with VEGFA when compared to VEGFA-alone treatment ( Fig 3B and 3C ) . Notably , the promoting effect of CCL2/VEGFA combination was stronger than the combination of CCL5/VEGFA and was comparable to that of CCL2/CCL5/VEGFA combination or CCL2 , CCL5 , IL-6 , IL-8 , and VEGFA combination , which was used as a positive control ( Fig 3B and 3C ) . Furthermore , depletion of VEGFA , CCL5/VEGFA , CCL2/VEGFA , or CCL2/CCL5/VEGFA by adding neutralizing antibodies into CAFs-LCCs co-culture system resulted in reduction of LCCs migration ( approximately 20% , 40% , 60% , and 63% ) and colony formation ( approximately 28% , 32% , 48% , and 52% ) , respectively , when compared to CM from CAFs and LCCs co-culture ( Fig 3D and 3E ) . These results indicated that CCL2/VEGFA combined plays a major role in the synergistic interaction between CAFs and LCCs , although CCL5 may also be involved in this process with a weaker effect when compared to CCL2 . Folkman et al . , reported that although VEGF can be released by cancer cells themselves , fibroblasts and inflammatory cells are the principal source of host-derived VEGF [21] . Fibroblasts are also the main source of CCL2 in response to tissue injury [22] , cytokine stimulation [23] , and cancer cell interaction [24] . Based on these findings , we hypothesize that the differences in cytokine secretion ( VEGF and CCL2 ) in the CM are caused by CAFs . To verify this hypothesis , we generated stable CAFs-shCCL2/VEGF or A549-shCCL2/VEGF double knock-down cell lines and CAFs-shScr or A549-shScr cell lines , as controls . We individually co-cultured CAFs-shCCL2/VEGF or A549-shCCL2/VEGF with either A549-shScr or CAFs-shScr . Only the depletion of CCL2 and VEGF expression in CAFs resulted in a great reduction of CCL2 and VEGF levels in the CM indicating that the up-regulation of CCL2 and VEGF in the co-culture was mainly secreted by CAFs ( Fig 3F ) . To investigate the molecular mechanisms of how miR-1 , miR-206 and miR-31 affect NFs-CAFs conversion , we searched several well-developed miRNAs algorithms and obtained a list of possible mRNA targets of miR-1 , miR-206 and miR-31 . Among the search results , CCL2 , VEGFA and FOXO3a captured our attention based on the following reasons . Fig 3A–3E results indicated that CCL2 and VEGFA play critical roles in CAFs and LCCs co-culture system . FOXO3a is known as a tumor suppressor which functions as a trigger for apoptosis [25] . CCL2 and VEGFA , and FOXO3a were predicted to contain seed matches for miR-1 , miR-206 and miR-31 , respectively ( Fig 4A ) . Since VEGFA has been reported as a target of miR-1 and miR-206 [26] , we focused on verifying CCL2 by miR-1 and miR-206 and FOXO3a by miR-31 . Luciferase reporter constructs were made to contain the putative binding sites of CCL2 or FOXO3a 3’-UTR wild-type regions ( WT ) , or with 3 nucleotide substitutes in their 3’-UTR regions ( Mut ) . Co-transfection of CCL2-WT and CCL2-Mut constructs with miR-1 , miR-206 or scramble miRNA into 293T cells showed that both miR-1 and miR-206 suppressed CCL2 wild type , but not mutant reporter activities ( Fig 4B ) . Furthermore , over-expression of miR-1 , miR-206 decreased CCL2 and VEGFA levels; whereas down-regulation of miR-1 , miR-206 increased CCL2 and VEGFA levels in CAFs ( Fig 4C ) . Similar effects of miR-31 on targeting FOXO3a were observed ( Fig 4D ) . Taken together , these results suggested that CCL2/ VEGFA and FOXO3a are direct targets of miR-1 , miR-206 and miR-31 , respectively . FOXO3a negatively regulates VEGFA expression at transcriptional level [27] . Indeed over-expression of FOXO3a decreased VEGFA levels but not affect CCL2 levels ( Fig 4E ) . Furthermore , the expression levels of VEGFA , but not CCL2 , were increased in CM when miR-31 was overexpressed in NFs , and were attenuated with miR-31-knockdown CAFs ( S7 Fig ) . These results suggested that miR-31 indirectly affects VEGFA expression in co-cultures via targeting FOXO3a . FOXO3a over-expression did not significantly affect CAFs cell apoptosis and cell proliferation ( S8A–S8D Fig ) . Over-expression of FOXO3a in CAFs did not change LCCs migration , but significantly impaired LCCs colony formation and this inhibition effect can be rescued by the addition of VEGFA ( S9A and S9B Fig ) . These results suggested that FOXO3a modulates VEGFA expression in fibroblasts to affect tumor microenvironment . To test whether CCL2 and VEGFA are sufficient targets of miR-1 , miR-206 and miR-31-reprogrammed NFs-CAFs conversion , we performed CCL2/VEGFA loss- and gain-of-function experiments in co-culture . As shown in Fig 5A and 5B , the NFs-TM-promoted LCCs migration and colony formation effect was blocked by the addition of anti-CCL2/VEGFA in co-culture , whereas CAFs-TM-induced the inhibition effect on LCCs migration and colony formation can be rescued by the additional of CCL2 and VEGFA in co-culture . Furthermore , over-expression of CCL2 and VEGFA in NFs significantly enhanced LCCs migration and colony formation , which could be abolished by the addition of anti-CCL2/VEGFA in co-culture . Similarly , applying recombinant CCL2 and VEGFA in co-culture were sufficient to restore the effect of si-CCL2/VEGFA-inhibited LCCs migration and colony formation ( S10A and S10B Fig ) . These results strongly suggest that miR-1 , miR-206 and miR-31 reprogram NFs-CAFs conversion via affecting CCL2/VEGFA expression . Antibody-based therapy has been widely used to treat cancer and miRNAs have emerged as therapeutic options to treat cancer in recent years [28–30] . To investigate the therapeutic effect of CCL2- and/or VEGFA-neutralizing antibodies or single- or triple-miRNAs delivery on lung tumor growth , we applied injection of CCL2- and/or VEGFA-neutralizing antibodies or formulated miRNAs . We administered CCL2- and/or VEGFA-neutralizing antibodies by intraperitoneal ( i . p . ) injections or formulated miRNAs by intravenous ( i . v . ) tail vein injection to BALB/c athymic nude mice bearing pre-established lung tumors . As shown in Fig 5C–5F , i . p . delivery of anti-CCL2 , anti-VEGFA and anti-CCL2/VEGFA combination dramatically reduced lung tumor growth , angiogenesis , TAMs accumulation , and tumor metastasis in mice . More importantly , the anti-CCL2/VEGFA combination exerted a synergetic effect as compared to anti-CCL2 or anti-VEGFA alone . Similarly , triple- and double-miRNA delivery resulted in a significant reduction of tumor burden , angiogenesis , TAMs accumulation , and tumor metastasis compared to animals receiving single miRNAs ( Fig 5G–5J ) . The up- and down- regulation of miR-1 , miR-206 and miR-31 in the tumor microenvironment ( mixture of fibroblasts and tumor cells ) were confirmed by Taqman qRT-PCR ( S11 Fig ) . Interestingly , the expression of CCL2 and VEGFA was markedly lower in triple- and double-miRNA delivery tumors compared with that in single miRNA tumors ( S12 Fig ) .
The cancer-promoting role of CAFs is unambiguously established [31] . CAFs are responsible for the synthesis , deposition and remodeling of ECM in tumor stroma , and also secrete paracrine growth factors that influence the growth of carcinoma [32 , 33] . However , it is largely unknown how quiescent fibroblasts are transformed into CAFs . To address this question , we compared miRNAs expression profiles between CAFs and NFs from the same lung cancer patients and applied fibroblast-cancer cell co-culture system to verify these findings in vitro . Our results suggested that altering the expression of three miRNAs ( miR-1 , miR-206 and miR-31 ) contributes to NFs converting into CAFs . Down-regulation of miR-1 and miR-206 and up-regulation of miR-31 in NFs ( NFs-TM ) promoted lung cancer cell migration and colony formation in vitro and tumor growth and lung metastasis in vivo when compared to NFs-Scr , and reversing the expression of these miRNAs in CAFs ( CAFs-TM ) blocked the enhanced LCCs migration , colony formation and tumor growth seen when using CAFs-Scr , providing evidence that miRNAs play an important role in reprogramming NFs into CAF-like fibroblasts that possess tumor-promoting functions . We note , however , that CAFs-TM only partially but not statistical significantly inhibit tumor lung metastasis and TAMs recruitment in the primary site of tumors compared to CAFs-Scr . One potential explanation is that the expression levels of CCL5 , an important chemokine for cancer cells metastasis and macrophage infiltration [14 , 34 , 35] , were not affected in CAFs-TM compared to CAFs-Scr . Thus , residual levels of CCL5 may partially contribute to this phenomenon . Comparing the cytokine expression profile of conditioned media from NFs-Scr-A549 , CAFs- Scr-A549 and NFs-TM-A549 , we identified the top five up-regulated cytokines ( CCL2 , CCL5 , IL-6 , IL-8 , and VEGFA ) that are commonly up-regulated in CM from CAFs- Scr-A549 and NFs-TM-A549 when compared with those from NFs-Scr-A549 . By gain- and loss-of-function studies , we identified the combination of CCL2/VEGFA to play a predominant role in the promotion of LCCs migration and colony formation in CAFs-LCCs co-culture system . More importantly , we further demonstrated a novel link between CCL2/VEGFA up-regulation and miR-1 , miR-206 suppression in co-cultured CAFs suggesting that CCL2 and VEGFA are directly targeted by miR-1 , miR-206 through binding to their 3’-UTR regions . CCL2 has been reported to function as a chemo-attractant [24] , and VEGFA is an important cytokine involved in angiogenesis and tumor growth [36] . Both are known to play important roles in progression and malignancy in multiple cancers , including breast cancer , prostate cancer , ovarian cancer , and lung cancer [37–39] . Consistent with these findings , our results clearly show that CCL2 and VEGFA double-knockdown in CAFs decreases , whereas CCL2 and VEGFA over-expression in NFs increases LCCs migration and colony formation , indicating the critical roles of CCL2 and VEGFA in NFs-CAFs transformation . CCL2 also plays a crucial role in the recruitment of inflammatory macrophages to the tumor site and become TAMs that are suggested to enhance tumor malignancy [40] . Interestingly , CCL2 overexpression induces tumor angiogenesis via TAMs [41] , and VEGFA production in regions of hypoxia in growing tumors benefits TAM accumulation [42] . Our results showed that tumors generated from CAFs- and NFs-TM- A549 commingled produced more pro-cancerous secreted factors than those generated from A549 cells alone , NFs- and CAFs-TM- commingled lung tumor . The VEGFA and CCL2 mainly released by fibroblasts in the tumor microenvironment are responsible for tumor angiogenesis and TAMs accumulation , thereby boosting tumor growth and metastasis . Furthermore , we found that FOXO3a , a well-known tumor suppressor , was down-regulated in CAFs compared with NFs . We further identified FOXO3a as a target of miR-31 and that up-regulation of miR-31 may contribute to the down-regulation of FOXO3a in CAFs . FOXO3a functions as a tumor suppressor gene mainly through triggering cell cycle arrest , repair of damaged DNA , and apoptosis [43 , 44] . However , over-expression of FOXO3a did not significantly induce CAFs apoptosis and cell growth inhibition . The reason may be due to the relatively low growth rate of fibroblasts compared to cancer cells . Besides , fibroblasts cell cycle progression is also highly controlled by PTEN , which is commonly mutated in cancer cells [45] . Considering that fibroblasts affect the function of cancer cells via paracrine , we suggested that FOXO3a functions as a tumor suppressor in fibroblasts and cancer cell co-culture through inhibition of VEGFA expression . We used two approaches to evaluate the potential therapeutic application of our current findings . One approach is to use a VEGFA neutralizing antibody injection . Interestingly , bevacizunb ( a neutralizing antibody against VEGFA ) has entered into phase II clinical trials for head and neck squamous cell carcinoma treatment [46] . We found that the combination of anti-VEGFA and anti-CCL2 yielded better suppression efficiency on tumor growth , angiogenesis , TAMs accumulation , and lung metastasis than either single antibody application . The other approach is using formulated miRNAs injection . MiRNA-based cancer treatments possess a number of advantages . MiRNAs can regulate a broad set of genes simultaneously and can modulate the tumor microenvironment affecting tumor cells and stroma . MiRNAs have shown reduced immune response and low toxicity when compared with lentivirus- or protein-based gene therapy [47–49] . Our results clearly showed that the systemic delivery of anti-miR-31 and miR-1 , miR-206 achieved a stronger anti-tumor progression effect than miR-1 , miR-206 or anti-miR-31 treatment alone ( Fig 6 ) . These results suggest that combination strategy is a promising treatment of lung cancer in the future .
FOXO3a ( sc-11351 ) and β-actin ( sc-1616 ) were obtained from Santa Cruz Biotechnology ( Santa Cruz , CA , USA ) . Fibronectin ( ab23750 ) , vimentin ( ab92547 ) , and α-SMA ( ab7817 ) were purchased from Abcam ( Massachusetts , US ) . CD31 ( 550274 ) were from BD Biosciences ( San Jose , CA , USA ) . Anti-Mouse CSF1R-PE ( AFS98 ) and Anti-Mouse F4/80 Antigen-APC ( BM8 ) were from eBioscience ( San Diego , CA , USA ) . CCL2 ( 279-MC ) , CCL5 ( 278-RN ) , IL-6 ( 206-IL ) , IL-8 ( 208-IL ) , and VEGFA ( 293-VE ) were purchased from R&D systems ( Minneapolis , MN , USA ) . HEK293T , A549 and H460 cells were purchased from ATCC ( American Type Culture Collection , Manassas , VA , USA ) . HEK 293T , A549 and H460 cells were maintained in a medium of RPMI 1640 medium supplemented with 10% FBS and 1% penicillin/streptomycin . NFs and CAFs were isolated as previously described [15] . We isolated CAFs from 15 human lung carcinomas and their corresponding counterpart NFs from their matched non-malignant adjacent tissues , taken at least 10 cm from the outer tumor margin . Both CAFs and NFs expressed fibroblastic markers , such as fibronectin and vimentin . The expression of α-SMA were much stronger in CAFs than corresponding counterpart NFs . NFs and CAFs were maintained in 1:1 mixture of DMEM and F12 medium supplemented with 400 ng/ml hydrocortisone , 200 ng/ml insulin , 15% FBS and 1% penicillin/streptomycin . Cells were maintained in 5% CO2 incubator at 37°C . A total of 15 paired fresh-frozen surgically resected lung tumors ( 8 adenocarcinomas , 6 squamous cell carcinomas , 1 larger cell carcinomas ) and matched non-malignant adjacent tissues were obtained from the first affiliated hospital of Nanjing Medical University . The study protocol has been approved by Ethics Committee of the First Affiliated Hospital and Nanjing Medical University ( approval No . 2015-SR-041 ) . Participating subjects provided written informed consent Lung carcinoma samples and normal tissues were confirmed by a pathologist . Peripheral blood ( 10 ml ) was drawn from each subject using standardized phlebotomy procedures . EDTA-Blood samples were collected without anti-coagulant and placed into two 5-ml red top vacutainers . After blood coagulation , serum was separated by centrifugation . All specimens were immediately frozen and stored at −80°C . For miRNAs levels analysis , 25 fmol of spiked-in cel-miR-39 ( Applied Biosystems ) was added in each plasma sample as an external control to monitor the quality of RNA extraction and normalization analysis . Cells were washed and lysed in RIPA lysis buffer with protease inhibitors ( Thermo Scientific ) . The total proteins were separated by 8 or 10% gradients SDS-PAGE gels . Proteins were transferred to a polyvinylidenedifluoride membrane and blocked with 5% nonfat milk . Then the membrane was incubated overnight with primary antibodies . Protein bands were detected by incubation with horseradish peroxidase ( HRP ) -conjugated antibodies and visualized with an enhanced chemiluminesence reagent . The expression levels of miRNAs were analyzed using Taqman MicroRNA Assay Kits ( Applied Biosystems , Foster City , CA ) specific for hsa-miR-1 , hsa-miR-206 and hsa-miR-31 . Expression of RNU6B ( U6 small nuclear RNA ) was used as an endogenous control . Normalization strategy for analysis of serum levels of hsa-miR-1 , hsa-miR-206 and hsa-miR-31 were previously described [50 , 51] . Briefly , the raw CT data for plasma miRNAs were first normalized using the CT RNU6B and then scaled to the spiked-in cel-miR-39 to correct for differences in extraction efficiency . To determine the quantity of VEGF , CCL2 , MMP9 , CCL5 , and SDF-1 mRNA , the cDNA was amplified by real-time PCR with Power SYBR Green PCR Master Mix ( Applied Biosystems ) , and the housekeeping gene GAPDH was used as the internal control . A relative fold change in expression of the target gene transcript was determined using the comparative cycle threshold method ( 2−ΔΔCT ) . All experiments were performed in triplicate . All primers used are listed in S1 Table . Microarray hybridization was performed as previously described [52] . Briefly , the miRCURY LNATM ( Locked Nucleic Acid ) microRNA version 11 . 0 microarray ( Exiqon , Denmark , http://www . exiqon . com/microrna-microarray-analysis ) was used , and each probe was repeated 4 times in the microarray . MiRNAs were labelled with Hy3TM or Hy5TM fluorescent groups using the miRCURYTM Array Power Labeling reagent kit to form fluorescent probes . The background was removed from the signal value , and scale normalization was done . The ratio between groups > 1 . 5 times or < 0 . 65 times , and a P value < 0 . 05 revealed by t test indicated the miRNAs were differentially expressed . The ratio > 1 . 5 times was defined as up-regulation , while the ratio < 0 . 65 times was defined as down-regulation . CCL2 and FOXO3a 3’-UTRs containing predicted miR-1 , miR-206 and miR-31 binding sites and corresponding mutant sites were amplified by PCR from genomic DNA ( HEK293T cells ) using pfu DNA polymerase ( Stratagene , CA , USA ) , respectively . The PCR productions containing the wild-type or mutant putative target sites of the CCL2 and FOXO3a 3’-UTR regions were inserted into untranslated region ( UTR ) downstream of the luciferase gene in the pMIR-reporter luciferase vector ( Ambion ) . Cells were cotransfected using Lipofectamine ( Invitrogen ) with 200 ng of wild-type or mutant luciferase reporter plasmid , 100 ng of β-galactosidase ( β-gal ) plasmid , and 100 nmol of pre-miR-1 , pre-miR-206 , pre-miR-31 or negative control precursor . Luciferase activity was measured 48 hours after transfection using β-gal for normalization . Experiments were performed in triplicate in three independent experiments . All primers used are listed in S1 Table . Cell migration assays were performed using migration chambers . LCCs were co-cultured with CAFs or NFs in complete medium at a ratio 1:1 , in 24-well plates for 24 hours . LCCs cells were seeded into a Boyden chamber with serum-free medium . Cells that had not migrated were removed from the interior sides of the chamber by cotton swabs . The exterior sides were fixed with 100% cold methanol and stained with crystal violet . Cells were counted under a microscope . Tumor samples were fixed with Z-Fix solution ( Anatech LTD , MI , USA ) for 24 h and processed by the paraffin-embedded method . The tissues sections ( 5 mm thick ) were then heat-immobilized or pepsin-immobilized according to the manufacturer’s instructions . Antibodies against vimentin or CD31 were used for the immunostaining and detected through the Dako Envision two-step method of immunohistochemistry ( Carpinteria , CA , USA ) . The relative angiogenesis levels were calculated by microvessel density ( MVD ) as described [53] . In short , slides were first scanned under low power ( ×40 ) in order to determine three areas with the maximum number of microvessels that were consequently evaluated at × 200 magnifications . For immunofluorescence assay , samples were stained with fibronectin , vimentin , and α-SMA after blocking with bovine serum albumin . Samples were incubated with Alexa Fluor 568-conjugated secondary antibody ( red ) or Alexa Fluor 488-conjugated secondary antibody ( green ) ( Life technologies , Frederick , MD , USA ) . Microscopic observation was performed under a fluorescence microscope ( Zeiss , Thornwood , NY , USA ) . Cells were washed and resuspended in HEPES buffer containing recombinant PI and Annexin V-FITC ( BD Biosciences , San Jose , CA , USA ) . The stained cells were analyzed with flow cytometry . For colony formation assay , 1 ml of 0 . 5% SeaPlaque agarose ( BMA , ME , USA ) was added to each well of 6-well plates . After solidation , 5 × 103 cells were mixed with 2 ml of 0 . 5% SeaPlaque agarose and added onto the top of the well . The CM were added to the wells and be replaced by every two days . After 12 days of culture , colonies were fixed with 100% methanol for 15 min and stained with 0 . 1% crystal violet . Colonies with diameter more than 1 . 5 mm were counted . Tumor tissue was minced and digested with dissociation buffer ( 100 U/ml Collagenase type IV and 100 μg/ml DNase in RPMI + 10% FBS ) in a shaking incubator at 37°C for 30 min . Digested tissues were filtered through 70-μm cell strainers . Cells were incubated with Fc Block . To identify TAMs , cells were stained with CSF1R-PE and F4/80-APC at 4°C for 30 min . Unstained control and single-stained cells were prepared in every experiment for gating . Dead cells were gated out by side-scatter and forward-scatter analysis . The cells were seeded 24 hours , and transfected with 100 nM of miRNA precursors or 100 nM anti-miRNA inhibitors ( Ambion , TX , USA ) using Lipofectamine RNAiMAX ( Invitrogen , CA , USA ) according to the manufacturer’s instructions . The expression levels of miRNAs were verified by stem-loop qRT-PCR . Cells transfected with scramble oligonucleotides were used as a negative control . Lentiviruses were generated by transfection of HEK293T cells with transducing vector and packaging vectors . Virus particles in the medium were initially harvested after 24 h and harvested every 12 or 24 h thereafter . Collected virus particles were filtered and transduced into target cells . Cells underwent two rounds of selection with appropriate antibiotics . Lentiviral plasmids expressing shScr , shCCL2 , and shVEGFA were obtained from Sigma Aldrich ( St . Louis , MO , USA ) . Lentiviral plasmids expressing CCL2 , VEGFA and FOXO3a were obtained from GeneCopoeia ( Rockville , MD ) . The levels of cytokines , growth factors and chemokines in the culture media were assessed by Bio-PlexPro human cytokine , chemokine , and growth factor array ( Bio-Rad Life Sciences , CA , USA ) using a Luminex 100 plate reader ( Bio-Rad Life Sciences , CA , USA ) according to the manufacturer’s protocols . CCL2 and VEGFA protein levels in CM were measured by ELISA kits ( R&D systems , Minneapolis , MN , USA ) according to the manufacturer’s protocol . 4–6 week aged nude mice of the BALB/c strain were purchased from the Nanjing General Hospital of Nanjing Military Command ( Nanjing , China ) . All nude mice were raised in and all experiments were conducted under SPF-level barrier system . A549 tumor cells ( 1×106 ) were commingled with fibroblasts ( NFs or CAFs ) , and mixed with Growth Factor Reduced Matrigel ( BD Lifesciences ) and injected subcutaneously into the right flank of each animal . Primary tumors and lung tissues were harvested from mice after 6 weeks after injection . Lungs were paraffin-embedded and serial sections were histologically examined with hematoxylin and eosin ( H&E ) stain . For quantitation of lung tumor foci , tumor numbers of 5 serial sections per lung were counted and totaled . The lung metastasis index for each mouse was calculated as the ratio of the number of foci colonies observed in the lungs divided by the mass of the primary tumor ( in grams ) and normalized to WT as fold changes [14] . ( Mean ± SEM; n = 8 ) . For neutralizing antibodies ( R&D system ) treatment , mice received i . p injections of single or combination of anti-mouse CCL2 ( MAB-497 ) , anti-mouse VEGF164 ( AF-493 ) or mouse IgG isotype control ( MAB002 ) twice a week starting on day 7 after tumor cell implantation for up to 6 weeks ( 2 mg/kg/dose ) . For miRNAs injection treatment , miRNAs were formulated with MaxSuppressor in vivo RNALancerII ( Bioo Scientific , Austin , TX , USA ) according to the manufacturer’s instructions . Each does contained 20μg of formulated oligo , which equals 1 mg/kg per mouse with an average weight of 20g . Formulated miRNAs were intravenously ( i . v ) by tail vein injections every 5 days starting on day 5 after tumor cells implantation . Results were analyzed using the version 13 SPSS statistical software ( SPSS , Chicago , IL , USA ) . Quantitative variables were analyzed between two groups using Student's t-test or among multiple groups using one-way analysis of variance ( ANOVA ) . Differences were considered significant at p<0 . 05 . | During tumorigenesis , normal fibroblasts ( NFs ) within the tumor stroma acquire a modified phenotype and become cancer-associated fibroblasts ( CAFs ) . CAFs provide oncogenic signals to facilitate tumor initiation , progression , and metastasis . Here , we set out to determine the factors that mediate the conversion of NFs into CAFs , focusing on miRNAs and secreted factors . Down-regulation of miR-1 and miR-206 and upregulation of miR-31 were found in CAFs derived from human lung cancer compared to paired NFs . Dysregulation of miR-1 , miR-206 and miR-31 expression promotes the conversion of NFs into CAFs through regulating VEGFA , CCL2 and FOXO3a expression . In addition , down-regulation of miR-1 and miR-206 and up-regulation of miR-31 has been observed in lung cancer patient plasma . More importantly , we demonstrated that systemic delivery of anti-VEGFA/CCL2 or pre-miR-1 , pre-miR-206 and anti-miR-31 dramatically decreased tumor angiogenesis , TAMs accumulation , tumor growth and lung metastasis . In conclusion , our data showed that miRNAs-mediated FOXO3a/VEGF/CCL2 signaling plays a prominent role in transforming NFs into CAFs , thus providing further support for the development of new diagnostic and therapeutic approaches to lung cancer . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"cytokines",
"cardiovascular",
"physiology",
"gene",
"regulation",
"cancer",
"treatment",
"immunology",
"cancers",
"and",
"neoplasms",
"basic",
"cancer",
"research",
"fibroblasts",
"oncology",
"angiogenesis",
"developmental",
"biology",
"micrornas",
"tumor",
"physiology",
"connective",
"tissue",
"cells",
"molecular",
"development",
"secondary",
"lung",
"tumors",
"tumor",
"angiogenesis",
"animal",
"cells",
"lung",
"and",
"intrathoracic",
"tumors",
"gene",
"expression",
"connective",
"tissue",
"biological",
"tissue",
"immune",
"system",
"biochemistry",
"rna",
"metastasis",
"cell",
"biology",
"nucleic",
"acids",
"anatomy",
"physiology",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"non-coding",
"rna"
] | 2016 | Reprogramming of Normal Fibroblasts into Cancer-Associated Fibroblasts by miRNAs-Mediated CCL2/VEGFA Signaling |
Patterned spontaneous activity in the developing retina is necessary to drive synaptic refinement in the lateral geniculate nucleus ( LGN ) . Using perforated patch recordings from neurons in LGN slices during the period of eye segregation , we examine how such burst-based activity can instruct this refinement . Retinogeniculate synapses have a novel learning rule that depends on the latencies between pre- and postsynaptic bursts on the order of one second: coincident bursts produce long-lasting synaptic enhancement , whereas non-overlapping bursts produce mild synaptic weakening . It is consistent with “Hebbian” development thought to exist at this synapse , and we demonstrate computationally that such a rule can robustly use retinal waves to drive eye segregation and retinotopic refinement . Thus , by measuring plasticity induced by natural activity patterns , synaptic learning rules can be linked directly to their larger role in instructing the patterning of neural connectivity .
Though synaptic plasticity is a feature of most excitatory synapses in the brain , how it functions in realistic contexts is largely unclear because its effects usually only manifest on the system level . The synaptic refinement of retinal ganglion cell ( RGC ) axons in the developing lateral geniculate nucleus ( LGN ) during early postnatal development in rodents provides an opportunity to study synaptic plasticity in this larger context , since the activity over the inputs to the LGN and the resulting developmental outcome are both well characterized [1] . At the earliest ages studied , LGN neurons receive inputs from a large number of RGCs from both eyes , and this number reduces to one or a few over the course of development [2–5] . At the system level , this synaptic refinement results in segregation into eye-specific regions and establishment of fine retinotopy , with neighboring RGCs projecting to neighboring LGN neurons [6] . This synaptic refinement—as well as similar refinement in the developing visual cortex [7 , 8] and superior colliculus [9]—is known to require spontaneously generated activity in the developing retina [10 , 11] . This activity consists of correlated bursts of action potentials that spread across large regions of the retinal ganglion cell layer [12] . These retinal waves have distinct spatiotemporal properties that have been studied in detail through a variety of multi-electrode [13 , 14] and calcium imaging studies [12 , 15 , 16] . Some aspects of the retinal wave activity , such as coincidences of RGC activity over second-long time scales , specifically contain information that could instruct synaptic refinement [14 , 17] . However , it is not known whether developing retinogeniculate synapses actually use the information available from retinal wave activity [18] . At one extreme , retinal activity may just be permissive , such that it is required for RGC axons to recognize chemical markers or stimulate outgrowth , but provides no additional instructions for development [18 , 19] . However , several recent studies have manipulated retinal wave activity and shown that some aspects of its spatiotemporal patterning are necessary for synaptic refinement [20 , 21] , suggesting an instructive role for retinal waves . It is generally thought that such instruction manifests at individual synapses as a synaptic learning rule that translates specific patterns of pre- and postsynaptic activity that arise from retinal waves into long-lasting changes in synaptic strength , ultimately resulting in stabilization of correctly projecting synapses and elimination of incorrect connections [22] . Although many different forms of synaptic learning rules have been observed at synapses throughout the brain [23]—including at the retinogeniculate synapse [3 , 24]—it is unclear how any such rule would operate in the context of the complex spatiotemporal patterning of activity provided by retinal waves . Here , we report a novel learning rule measured at the retinogeniculate synapse that is based on the system-level patterning of neuronal activity generated by retinal waves in vivo . This learning rule demonstrates different amounts of long-term synaptic strengthening and weakening based on the timing of bursts over a seconds-long temporal window . It thus is qualitatively different from previously observed forms of synaptic plasticity , though we demonstrate how it is consistent with spike-time dependent plasticity [25] . Importantly , because burst-time–dependent plasticity ( BTDP ) describes the changes that synapses make in response to realistic activity , we demonstrate that it leads to the developmental refinement of the LGN observed on the system level and , furthermore , suggests how to predict the results of such alterations in activity [7 , 9 , 19–21 , 26 , 27] . In doing so , we present a full description of how natural activity patterns likely guide system-wide retinogeniculate refinement through changes in synaptic efficacy at the synapse level .
Using perforated patch recording from an in vitro slice preparation of the LGN and optic tract ( OT ) , the effects of the natural activity patterns produced by retinal waves on selected retinogeniculate synapses were examined . Throughout most of the period of eye segregation , RGCs provide the major source of driven input to the LGN [28 , 29]; as a result , the population-level imaging of retinal waves provides a full view of the spatiotemporal dynamics across the inputs to LGN neurons ( Figure 1A ) . In LGN neurons , these inputs manifest as large synaptic currents lasting seconds , and evoke bursts of action potentials in response [30] . Thus , to replicate the effects of retinal waves on the synapse , we combined minimal stimulation of the OT ( to activate one or a few synapses ) with direct current injection into the LGN neuron ( to simulate the remainder of the inputs ) . Such current injection was adjusted to evoke physiologically appropriate LGN activity: 10–20 Hz spiking for 1 s [30] . Though both the participation of individual RGCs in retinal waves [14] and the retinal waves themselves [16] are highly variable , previous analysis of the spatiotemporal properties of the retinal waves demonstrated that the information relevant for driving synaptic refinement is only contained in the relative timing of bursts , rather than other details of the bursts themselves [17] . Thus , to tie changes in the retinogeniculate synaptic efficacy to instructive features of the retinal wave activity , we measure how the strength of the selected synapses are affected by particular “burst latencies” between presynaptic OT stimulation ( 10 Hz for 1 s ) and postsynaptic depolarization ( Figure 1B ) . Stimulation of the synapses activated via OT stimulation has little effect on the firing of the postsynaptic neuron , due to the small amplitude of synaptic inputs , as shown in a comparison between the firing pattern of the neuron with ( Figure 1B , top ) and without ( Figure 1B , bottom ) OT stimulation . This is in marked contrast to that of tetanic stimulation , which dramatically alters postsynaptic activity patterns ( Figure 1C ) . Despite the small contributions of synaptic stimulation to the firing of the postsynaptic neuron , pairing pre- and postsynaptic bursts reliably leads to long-lasting increases in the efficacy of the synapse , as shown in an example in which presynaptic stimulation and postsynaptic depolarization were completely overlapping ( Figure 2A ) . The increase in the strength of the synapse resulting from this “zero latency” stimulation is usually gradual ( over the course of 15 min ) and modest ( 31 . 0% in this example ) . This time coarse is in marked contrast to that of long-term potentiation ( LTP ) induced by tetanic stimuli , which typically leads to an initial 2- to 3-fold increase in excitatory postsynaptic current ( EPSC ) size and then a long-lasting increase as high as 100% ( as in a previous study of plasticity in the LGN [24] ) . In contrast , the average magnitude of synaptic enhancement that we measured was 21 . 3 ± 5 . 1% ( n = 7 ) , and this increase in efficacy builds gradually over roughly 15 min ( Figure 2B ) . This modest enhancement of synaptic efficacy was only observable using perforated patch recording , and was not seen during standard whole-cell recording or following a patch rupture during a perforated patch recording . Furthermore , as a result of the small synaptic currents observed ( typically ~20 pA ) , small soma size ( mean membrane resistance of 925 MΩ ) , and high serial resistance ( RS = 50–100 MΩ ) typical of recordings from these very immature neurons , a large number of recordings contained changes in EPSC size that resulted from fluctuations in RS and/or patch rupture . As a result , strict criteria were applied to eliminate more than 90% of the recordings due to these variable factors ( see Materials and Methods ) . To determine whether the observed synaptic enhancement depends on coincidence of pre- and postsynaptic bursts or simply on the occurrence of presynaptic stimulation and/or postsynaptic depolarization , we performed the identical stimulation protocol described above with a latency of −1 , 100 ms; in this case , there is no overlap between pre- and postsynaptic activity because presynaptic stimulation ends 100 ms before postsynaptic depolarization begins . This “non-overlapping” latency caused a modest decrease in synaptic efficacy ( Figure 2C ) , which was consistent across n = 6 experiments performed at this latency ( −6 . 2 ± 1 . 7% ) . Such a decrease was consistent with a range of other non-overlapping latencies used , including +1 , 100-ms latency ( −3 . 5% , n = 3 ) , +1 , 200-ms latency ( −2 . 1% , n = 1 ) , +1 , 500-ms latency ( −10 . 9% , n = 1 ) , −2 , 100-ms latency ( −10 . 7% , n = 1 ) , and presynaptic stimulation without any postsynaptic depolarization ( −4 . 5% , n = 1 ) . The average of all non-overlapping burst cases is shown in Figure 2D , representing a statistically significant average decrease of −5 . 9 ± 1 . 4% in synaptic efficacy ( n = 13 , p < 0 . 01 ) . Together , these observations demonstrate that retinal wave activity can evoke either homosynaptic potentiation or depression depending on burst latency ( Figure 2E ) , and thus provides a mechanism for competition between inputs [31] . Burst latencies between zero ( fully overlapping ) and ±1 , 000 ms ( non-overlapping ) evoked intermediate levels of plasticity . A summary of 39 experiments in which changes in synaptic efficacy could be reliably gauged is shown in Figure 3 , and the results of an additional six experiments using 2-s pre- and postsynaptic bursts , which demonstrate consistent results , are shown in Figure S1 . This burst-time–dependent learning rule is a concrete instance of Hebbian plasticity , since synaptic efficacy increases for “cells that fire together , ” whereas synapses between neurons that are “out of sync” are weakened [22] . In fact , the tent-like shape of this learning rule demonstrates that—under the conditions studied—the total change in synaptic strength is related to the overlap between pre- and postsynaptic bursts , adjusted such that non-overlapping latencies result in synaptic weakening . Assuming that the overall burst latency , without regard to burst order , is linearly proportional to the observed changes in synaptic efficacy leads to the “symmetric” learning rule ( dashed line ) shown in Figure 3 . To investigate whether the order of pre- versus postsynaptic bursting affected the amount of plasticity , we allowed a different linear relationship for positive versus negative latencies ( solid line ) . However , the resulting “asymmetric” rule was not significantly different; thus it appears that burst order plays little role in determining synaptic changes . One of the most notable aspects of the observed burst-based learning rule is the second-long temporal window over which the magnitude of change in synaptic efficacy depends on the burst latency , and is in marked contrast to the much shorter temporal window ( ~10 ms ) of spike-timing–dependent plasticity ( STDP ) [25] . The orders-of-magnitude difference in time scale between STDP and BTDP likely arises from the time scale of the bursts themselves , since a given burst pairing is associated with an ensemble of shorter latencies between pre- and postsynaptic spikes; three examples are shown in Figure 4A ( corresponding to experiments numbered 1 , 2 , and 3 in Figure 3 ) . It is therefore possible that the observed burst-based rule is the cumulative effect of a spike-based rule applied to the individual spike latencies that comprise a given burst pairing . To discover whether a spike-based learning rule might underlie the observed BTDP , we evaluate the predictions of both spike-based and burst-based rules by calculating correlation coefficients ( CC ) between their predicted changes in efficacy and those observed in the 39 experiments of Figure 3 ( see Materials and Methods ) . As a baseline , the burst-based learning rules of Figure 3—which assume the amount of synaptic change is linearly related to burst latency—yield CC = 0 . 71 for the symmetric rule and CC = 0 . 73 for the asymmetric rule ( Figure 4B , first column ) . To compare , we apply the spike-time–dependent learning rules measured in the Xenopus retinotectal system [32] and rat somatosensory cortex [33] to the measured spike-latency distributions ( e . g . , as shown in Figure 4A ) . First , we assume that each spike latency contributes independently to the total change in synaptic efficacy such that the all pairs of pre- and postsynaptic spikes that occur during each plasticity experiment ( e . g . , those comprising the spike-latency distributions shown in Figure 4A ) can be used to generate a prediction of the total amount of synaptic change in each case . Doing so demonstrates that a naively applied STDP is a poor predictor of the observed plasticity , since the Xenopus rule has a correlation coefficient CC = 0 . 51 , whereas the cortex rule is substantially worse and predicts synaptic strengthening when weakening was observed and vice versa , resulting in CC = −0 . 42 ( Figure 4B , #1 ) . These predictions improve , however , when only nearest neighbor ( “close” ) spike pairs are considered and the other pairs are ignored ( Figure 4B , #2 ) , especially for the cortex rule in which the long time window for depression usually encompasses many spike pairs . In fact , previous studies of STDP in the context of more complicated neuronal activity [34 , 35] suggest phenomenological rules by which STDP can be modified to properly account for the observed synaptic plasticity . In particular , Sjöstrom et al . [34] suggest that , when individual spikes are associated with both positive and negative latency pairings , weakening is suppressed by strengthening . Applying this modified spike-based rule further improves the predictive power of both the Xenopus and cortex rules ( Figure 4B , #3 ) , though still not to the level of the burst-based rule . In particular , it leads to improvement by favoring strengthening over weakening for coincident bursts , which results in a net strengthening even though they have roughly equal numbers of positive ( depressing ) and negative ( potentiating ) spike pairs ( e . g . , see Figure 4A , middle ) . In this way , by ignoring the temporal window for depression , STDP predictions become increasingly consistent with the changes in synaptic efficacy observed in the context of the burst-based activity associated with retinal waves . This is consistent with the idea that , in the context of the burst-based activity present in the developing visual system , the effective learning rule predicted by STDP itself changes its form . We demonstrate this by plotting the average spike-based learning rule as a function of spike latency ( Figure 4C ) . First , in the naive application of STDP , every spike pair has its full effect , resulting in the normal STDP rules ( Figure 4C , left , dashed lines ) . However , when only nearest neighbor spike pairs are considered , it becomes increasingly likely that longer-latency pairs will be ignored , and as a result , the average change in synaptic efficacy at a given latency is reduced . Because the bursts used in these experiments involve spike rates around 10 Hz ( corresponding to roughly 100 ms in between each spike ) , the Xenopus rule is relatively unaffected by limiting the consideration to nearest neighbor spike pairs ( Figure 4C , top , solid line ) , whereas longer positive latencies are attenuated for the cortex rule ( Figure 4C , bottom , solid line ) . The Sjöstrom rule has a much great impact because it ignores positive-latency pairs that share a spike with negative-latency pairs , and results in a further attenuation of the window for synaptic weakening ( Figure 4C , dotted lines ) . Overall , This leads to a significant decrease in amount of synaptic weakening predicted by STDP in the context of natural retinal wave activity ( Figure 4C , right ) , especially for the cortex rule ( Figure 4C , bottom ) due to its long window of synaptic depression . Notably , the ability of a spike-based rule to predict the observed plasticity seems to be inversely related to the amount of synaptic weakening predicted by the rule . As a result , we evaluated a simple spike-based coincidence rule—with no temporal window for synaptic weakening—to compare with the previous STDP rules . If N is the total number of spike latencies within 50 ms , the best fit to data is given by Δ = 0 . 15N − 4 . 05 ( % ) , and has nearly the same predictive power ( CC = 0 . 68 ) as the burst-based rules of Figure 3 . The close correspondence of the predictions of this spike-based coincidence rule ( Figure 4D ) with the burst-based coincident rule ( dashed line ) suggests how BTDP—based on time scales on the order of a second ( arguably most relevant to the slow propagating activity in the retina ) —may be comprised of spike-based rules that act on shorter time scales that may be more appropriate for synaptic function . These considerations together show that synaptic plasticity of retinogeniculate synapses resulting from “natural” activity patterns are directly predictable from the total amount of coincidence between pre- and postsynaptic activity , whether considering either spikes or bursts . As a result , though a given LGN neuron may receive tens to hundreds of RGC inputs driven by the complex spatiotemporal properties of retinal waves [16] , synaptic development appears to be governed by a simple and robust computational principle: synaptic strengthening and weakening is proportional to the amount of coincident activity between RGC and LGN neurons . Does such “Hebbian” plasticity operate over time in a way that will drive retinal-wave–dependent development of the system ? Activity-dependent retinogeniculate refinement occurs over several weeks in most species , meaning that the gradual strengthening of some synapses and elimination of most others is a cumulative effect of many thousands of individual retinal waves . Figure 5 provides evidence that the effects of retinal waves are indeed cumulative . These two examples show sequential induction of opposing plasticity with stimulation protocols separated by an hour: in the first example , a long-lasting weakening is followed by strengthening ( Figure 5A ) ; the reverse occurs in the second example ( Figure 5B ) . These examples also provide another demonstration that the particular burst latency—rather than the mere presence of bursts—is responsible for determining the sign of plasticity . The cumulative effect of many retinal waves can be simulated across all the inputs to a given LGN neuron in a computer model of the retinogeniculate system ( see schematic , Figure 6A ) . We model this system using the burst-based learning rule in conjunction with two simulated retinas composed of two separate sets of roughly 19 , 000 RGCs that independently generate retinal wave activity using a simulation that accurately reproduces their experimentally described spatiotemporal properties [16 , 36] . A subset of these RGCs from both retinas are initially connected to a single LGN neuron . Because both the activity of the RGCs and that of the LGN neuron [30] are comprised of bursts that last over seconds , and the resulting burst-based learning rule is correspondingly coarse , the particular details of how the LGN activity results from the input RGC activity do not qualitatively affect the simulation results . Furthermore , the cumulative changes in synaptic efficacy are predictable simply from the total amount of coincident activity between a given RGC and LGN neuron . We simulate a short period of development ( 1 , 000 min ) in order to demonstrate how retinal wave activity , combined with BTDP , drives eye segregation and refinement of retinotopy . By limiting this simulation to this short period , we can demonstrate clear developmental trends without needing assumptions about how the system evolves over longer times , which involves changes in retinal wave properties [14] , functional changes in intra-LGN connectivity [37] , and other aspects of longer-term development that are not experimentally constrained . Consider an LGN neuron that initially receives input from a localized set of RGCs in each retina ( Figure 6A ) with an initial bias towards the left eye ( such that the right-eye connections are 20% weaker , but otherwise the same ) . The total amount of coincident activity between pre- and postsynaptic activity over the 1 , 000-min simulation is shown in Figure 6B as a contour plot over each retina ( top ) and also as a slice across the center of each retina ( bottom ) . Due to the initially stronger connection to the left eye , retinal waves in the left eye drive more postsynaptic activity and , as a result , RGCs in the left eye will have , on average , more coincident activity . As a result , a dotted line representing a balance of strengthening and weakening ( Figure 6B ) —no matter its exact location—will cut through a relatively higher section of the right-eye curve , demonstrating that a disproportionate number of the right-eye connections will become weakened ( shaded areas ) , and correspondingly more of the left-eye connections will become strengthened . Furthermore , notice that the shaded areas in both eyes ( Figure 6B , bottom ) are furthest from the center of the retinal area that the LGN neuron is connected to . This occurs because RGCs at the center will most likely be involved in a retinal wave that evokes postsynaptic activity and thus have the most coincident activity . Thus , this simulation of a small segment of development demonstrates the longer-term trends of RGC refinement—driven by the observed BTDP—to become increasingly retinotopic and eye segregated . These trends arise simply because retinal wave activity correlates local regions of RGCs to become active together and thus cooperatively drive postsynaptic activity . Note that such development simply reinforces existing biases that are likely established through activity-independent mechanisms [38 , 39]: the tight correspondence between initial bias and amount of coincident activity is shown in Figure 6C ( solid line ) . The amount of area in the LGN occupied by RGC axons from each eye can also be influenced by differences in the activity between the eyes [20] . The retinal wave model can also simulate the aberrant retinal wave activity that is induced by raising intracellular cyclic adenosine monophosphate ( cAMP ) levels through the application of adenosine [36] , which generates waves that are much more frequent and larger . We first simulated this aberrant activity in the left eye while keeping activity in the right eye normal , meaning that a given RGC in the left eye was involved in a retinal wave 2–3 times a minute ( instead of once every 2–7 min ) , and more LGN activity is evoked during a given left-eye wave because of its larger size . This results in a larger amount of coincident activity in the left eye for a wide range of bias in the initial connection strength ( Figure 6C , dashed line ) . These simulations thus reproduce the experimentally observed increase in territory of the more active eye [40] . Furthermore , when raised cAMP levels are simulated in both eyes , the amount of overlap again balances out ( Figure 6C , dotted line ) , resulting in normal eye segregation as experimentally observed [20] . However , due to the large size of the waves in the cAMP condition , retinotopic refinement is much more coarse ( Figure 6D ) , and as a result , may result in deficiencies in retinotopy . Thus , this simple model—based on the observed synaptic learning rule in the LGN—illustrates both the robustness of eye segregation and retinotopic refinement driven by retinal waves , and also provides a larger framework to understand the many recent experiments manipulating retinal wave activity that have different outcomes for patterns of retinogeniculate connectivity .
Detailed studies of both retinogeniculate refinement and the retinal activity that drives it provide a unique opportunity to relate rules governing synaptic plasticity to their role in guiding activity-dependent development . Guided by this knowledge , we reproduced the relevant aspects of the population activity in an in vitro preparation of the LGN and OT . We discovered a novel learning rule based on the relative timing between bursts of action potentials in the presynaptic population of RGCs and the postsynaptic target LGN neurons , such that short latencies cause potentiation and longer ones cause depression . The changes in synaptic efficacy evoked by these “realistic” stimulation patterns have a gradual onset ( Figure 2 ) and much smaller magnitude than a majority of synaptic plasticity observed with other stimulation protocols such as tetanus [24] , and thus their effects are likely gradual and cumulative over the course of development ( Figure 5 ) . One of the notable features of this observed BTDP is its seconds-long temporal window , which likely arises from the burst-based nature of the activity in the visual system at this stage of development [13 , 30] . Due to the compound aspect of this activity , we investigated whether it might arise from a shorter–time-scale rule such as STDP [25 , 32 , 33] . Though naively applied STDP cannot explain the observed plasticity , we found that modifications to STDP that account for multiple-spike interaction [34 , 35] result in much better predictions of the observed burst-based plasticity . Furthermore , the bursts present in retinal waves involve a much higher degree of multiple-spike interaction than has been studied [34 , 35] , and in this sense , it is likely that the simple spike-based coincidence rule that best explains the observed plasticity ( Figure 4C ) may be an extreme form of modified STDP . However , since the form of STDP induced by isolated spikes is not known in this system , whether or not a spike-based coincidence rule is a modified form of a more traditional STDP rule ( such as those measured in other systems , e . g . , Figure 4 ) is a matter for future investigation . Of course , the two examples of STDP considered in Figure 4 were observed in systems in which the action of single spikes is much more relevant: the “cortex STDP rule” is measured in the somatosensory cortex where single isolated spikes drive layer II/III neurons [41] , and the Xenopus retinotectal STDP [32] is present at a time when vision , rather than retinal waves , drives RGC activity . Likewise , evidence suggests that the mammalian visual cortex is governed by STDP at later ages when activity is driven by vision [42] , which results in less temporally and spatially correlated patterning of neuronal activity . In this way , the observations presented in this paper support how the relevant form of synaptic learning rules depends on the natural activity patterns that exist in a particular system . The burst-based learning rule that we observe also resembles other forms of previously observed synaptic plasticity , and is strikingly similar to pairing protocols that induce LTP in other systems [43] , because large synaptic currents that drive LGN neurons during retinal waves induce seconds-long depolarization [30] , and we here show that the observed burst-based learning rule is well predicted by the number of presynaptic spikes paired with this depolarization ( Figure 4C ) . In fact , tetanus-based plasticity also essentially amounts to depolarization paired with presynaptic stimulation ( Figure 1C ) , though the magnitude and frequency of this stimulation makes it unclear how it can apply to natural activity patterns . The ability to measure a learning rule that can be applied directly to known natural activity in the retinogeniculate system provides a framework for understanding a number of innovative recent experiments that have examined the link between retinal wave activity and patterning of RGC connections by disrupting natural retinal waves either pharmacologically or genetically [19–21 , 26 , 27] . Our observation that both homosynaptic strengthening and weakening are induced by natural activity patterns provides a mechanism for competition [31] that is necessary to understand the results of Stellwagen and Shatz [20] ( Figure 6 ) . In the meantime , there has been disagreement in the interpretation of experiments in which manipulations that disrupt retinal wave activity either prevented [11 , 21] or failed to prevent [19 , 21] eye segregation . Torborg et al . [21] concluded that high-frequency synchronized bursting between neighboring RGCs is necessary to drive eye segregation , whereas asynchronous spiking does not disrupt eye segregation . These results are entirely consistent with BTDP: high-frequency bursting among neighboring RGCs would elicit postsynaptic activity in connected LGN neurons and strengthen existing connection biases , whereas asynchronous spiking would not evoke postsynaptic activity . However , our model suggests that the crucial factor determining whether RGC axons from each eye segregate in the LGN is not activity between pairs , but the summed activity over local regions of the retina that would drive LGN activity ( which should be observable in the retina with multi-electrode or imaging ) . In this context , disrupted eye segregation might only be observed in conditions where RGC activity cannot consistently elicit postsynaptic spiking . The observed burst-time–dependent plasticity provides a simple computational principle for organizing an immature system driven by spontaneous activity; in fact a burst-based learning rule would be “safer” at developing synapses , where vesicle release is slow and more uncertain [44] . Furthermore , the simplicity and robustness of this rule suggests that it may exist at other developing visual synapses driven by retinal waves that have been shown to require retinal wave activity to drive refinement , including the visual cortex [7 , 8] and superior colliculus [9] . Such a Hebbian learning rule has been predicted at this synapse because the first observations of retinal waves [13 , 14 , 22] , and several modeling studies of the retinogeniculate system [45–47] , have shown that a Hebbian rule could use retinal wave activity to instruct refinement . In fact , a learning rule of this nature makes effective use of the information provided by the detailed spatiotemporal properties of the retinal waves [1 , 17] . Such a match between the information conveyed by retinal waves and the properties of the observed plasticity suggests a system-level organizational strategy in which the properties of retinal waves and activity-dependent plasticity are tuned to each other in order to drive and maintain activity-dependent development robustly .
Brain slices containing the LGN and OT were prepared from rat pups ( P7–11 ) . They were 350 μm thick , and cut in a tipped parasagittal plane [48] . The bathing solution ( ACSF ) contained ( in mM ) : 130 NaCl , 3 KCl , 1 . 25 KH2PO4 , 20 NaHCO3 , 10 glucose , 1 . 3 MgSO4 , 2 . 5 CaCl2 ( pH 7 . 4 , equilibrated with 95%O2–5% CO2 ) . Slices were held in a recording chamber on a fixed stage microscope and superfused ( 3–5 ml/min ) with ACSF at room temperature ( ~25 °C ) . Though no inhibitory postsynaptic currents ( IPSCs ) were observed during recording for ages under P12 despite using a low-chloride internal solution ( see below ) , picrotoxin ( 100 mM ) was added in a subset of experiments to the ACSF to block fast GABAergic transmission . Perforated patch current and voltage clamp recordings were made with electrodes pulled from borosilicate glass with a final resistance of 3–7 MΩ . The recording electrodes contained ( in mM ) : 100 K-gluconate , 4 NaCl , 20 KCl , 0 . 2 CaCl2 , 10 HEPES ( free acid ) , 1 . 1 ethylene glycol-bis ( b-aminoethyl ether ) -N , N , N′ , N′-tetraacetic acid ( EGTA ) , 2 Mg-ATP , 1 MgCl2 and 5 glutathione ( pH 7 . 2 ) , 50 mg/ml amphotericin , and ( for some experiments ) Lucifer Yellow ( K-salt , ~1 mg/ml; Molecular Probes , Eugene , Oregon , United States ) . Recordings were made with either an Axopatch 200B or 700A amplifier ( Axon Instruments/Molecular Devices , Sunnyvale , California , United States ) , digitized using a 16-bit A/D converter at 5–10 kHz and filtered at 2 kHz . All voltages were adjusted for an estimated electrode-bath junction potential of −12 mV by offline subtraction . Bridge correction was performed offline . Digitized data were analyzed by MATLAB ( Mathworks , Natick , Massachusetts , United States ) . Recordings were made from LGN neurons identified using differential interference contrast ( DIC ) visualization . Once a seal was obtained ( >1 GΩ ) , access resistance was monitored as it decreased ( due to the amphotericin ) and stabilized from 30–100 MΩ , at which point the synaptic plasticity experiments ( described below ) were commenced . Since a large problem in measuring plasticity was an accidental rupture of the patch during the experiment , several methods were used to control for this . Initially , Lucifer Yellow was included in the internal solution and was imaged using fluorescence microscopy every 30 s to verify that the membrane patch did not rupture ( which would result in the neuron filling ) . Recent experiments used an electrode brace attached directly to the headstage that increased electrode stability and removed most fluctuations in serial resistance , making a patch rupture identifiable from jumps in the access resistance alone . A bipolar stimulating electrode was placed into the OT . Stimulus current was turned down until it did not elicit input to the LGN neuron recorded using perforated patch , and then gradually turned up to until a stable EPSC was seen ( 7–140 pA ) . This “minimal stimulation” arguably stimulates one or few synapses , but in many cases the amount of input varied continuously with stimulus current , in which case the current was adjusted to where the EPSC size was relatively stable . In every case , the EPSC represented a small fraction of the total input to the LGN neuron , because turning up the OT stimulus current could evoke significantly larger currents , which could be a nanoamp or greater . The EPSC amplitude was monitored in an LGN neuron in voltage clamp every 30 s until a stable baseline was achieved . The recording mode was then switched to current clamp , and 1-s current steps of increasing amplitude were delivered until the neuron was driven to fire 10–20 spikes in the second interval . Then , 10 Hz OT stimulation was paired with this current injection at a prescribed latency , and this was repeated ten times at 40-sec intervals ( Figure 1 ) . The recording mode was then switched back into voltage clamp , and the resulting EPSC amplitude was monitored as it had been during the baseline recording . Serial resistance was constantly monitored over the experiment , and trials in which there was significant change ( denoting patch rupture ) or in which smaller changes in RS were correlated with changes in EPSC size were discarded . A given stimulation protocol evokes excitatory postsynaptic potentials ( EPSPs ) at precisely determined times {tE} determined by the times of OT stimulation adjusted by the synaptic latency determined during baseline recording . In the meantime , the timing of evoked LGN spikes {tspk} are directly observable from the current clamp recordings during the stimulation protocol . Together , these timings were used to form histograms of the latencies {L = tE − tspk} of three different classes of spike pairs: ( 1 ) comparisons of timings between all EPSPs and spikes; ( 2 ) comparison between nearest-neighbor spikes only; and ( 3 ) comparison between selected pairs of nearest neighbor spikes as suggested by Sjöstrom et al . [34] , which ignore pairs with positive latencies when either of the spikes in the pair was also paired with other spikes with a negative latency . These spike latency histograms—one set for each experiment—were then used to derive spike-time–dependent learning rule predictions based on a wide class of STDP-type rules . In particular , for a particular set of latencies {Li} , the predicted change in synaptic efficacy ΔP for a general spike-time-based rule is given by: Note that the constants A and C adjust the overall scaling and offset of the rule such that the resulting plasticity ΔP can be matched to the experimental data of Figure 3 . A and C are chosen to obtain the best fits for each learning rule , and do not affect the CC . Note that C can be negative , implying that that non-overlapping bursts—which may not be associated with any spike latencies in the relevant window—would result in synaptic weakening given by C . In our results , we report results from rules measured in the retinotectal system ( B = −1 , τ− = τ+ = 10 ms [32] ) ; and somatosensory cortex ( B = −0 . 3 , τ− = 10 ms , τ+ = 60 ms [33] ) . Local variations of the parameters of these rules were also investigated and did not yield significantly better fits . We also applied the rules for positive values for B in each case , as well as a rule that uses only the set of latencies NL within 50 ms . The best rule in this case was Δ = 0 . 145NL − 4 . 0472 ( % ) . We evaluated the fits of the learning rules using two methods: CC and mean-squared error ( MSE ) . The CC compared the observed plasticity data for all 39 experiments ( j = 1 to 39 ) {Δ0j} to the predicted plasticity {ΔPj} using CC = ∑j [Δ0j − mean ( Δ0 ) ] [ΔPj − mean ( ΔP ) ]/[std ( Δ0 ) std ( ΔP ) ] . A CC of 1 means that the data are perfectly predicted by the learning rule , and zero means that it is completely uncorrelated . The MSE is given by MSE = ∑j [Δ0j − ΔPj]2 . Because of the correspondence of the results of these two methods and the lack of dependence of the CC on the parameters A and C , we reported only CC results here . We also used the CC and MSE to determine the best burst-based learning rule . The best symmetric rule as a function of burst latency LB is given by Δ = 18 . 2 − 25 . 8 LB ( for |LB| < 1 ) , and Δ = −7 . 6 for |LB| ≥ 1 . The best asymmetric rule used 18 . 3% for zero intercept and −28 . 9 and −21 . 0 ( %/sec ) for the left and right slopes , respectively . A model of two retinas and a single LGN neuron were simulated as described in the Results ( Figure 6A ) . To generate retinal waves in each retina , we implemented the retinal wave model of Butts et al . [36] , which ran for 1 , 000 simulated minutes using two different random number seeds to simulate uncorrelated retinal wave activity in each eye . The weight matrices wL , i and wR , i that determine the initial strength of connection between the LGN neuron and each RGC in both eyes were given by two-dimensional Gaussian distributions centered at the center of each retina with a standard deviation of ten times the RGC cell spacing , and their overall magnitude was multiplicatively scaled to establish an initial bias . LGN activity at every time step R ( t ) was given by: where rL , i ( t ) and rR , i ( t ) are either 0 or 1 depending on whether the given RGC is bursting or not at time t . Note that particular physiological details of the LGN neuron only affect its activity over shorter time scales , and thus this general model for activity is sufficiently representative of a large range of conditions for the purposes of this simulation . The amount of coincident activity between a given RGC and the LGN neuron is then given by OL , R = ∑t R ( t ) rL , R , i ( t ) , and is plotted in Figure 4 for various simulations . The “resulting bias in coincident activity” ( Figure 6C ) is on a scale of zero ( right eye ) to one ( left eye ) and is given by OL/ ( OL+OR ) . Note that the reported results occur robustly for a large range of initial connectivities , and manipulating the few assumptions of this model do not qualitatively affect our results . The manipulated waves simulating conditions of increasing cAMP [40] represented a different parameter regime of the same retinal wave model [36] , and specifically , we used the spontaneous rate p = 0 . 002 and the threshold θ = 2 . 2 instead of the normal p = 0 . 030 and θ = 3 . 5 . All simulation code is available at the Web site http://rd . plos . org/10 . 1371_journal . pbio . 0050061_01 . | The brain is comprised of an immense number of connections between neurons , and clever strategies are required to achieve the correct wiring during development . One common strategy uses neural activity itself as feedback to instruct individual connections ( synapses ) through synaptic learning rules that delineate which patterns of activity strengthen the synapse and which weaken it . Throughout life , such activity-dependent synaptic changes are likely driven by experience and are thought to underlie learning and memory , but during early stages of development , they are often driven by activity spontaneously generated within the brain . Here , we study connections in the visual pathway between the retina and lateral geniculate nucleus ( LGN ) , which—to develop correctly—require spontaneous “retinal waves” before the eye is responsive to light . By replaying the retinal wave activity as it appears at single LGN synapses , we observe a novel learning rule that describes a relatively simple computation for the developing synapse in the context of retinal wave activity . We then demonstrate how this learning rule is matched to properties of the retinal waves in order to robustly drive the synaptic refinement that occurs in the visual system . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"developmental",
"biology",
"mammals",
"physiology",
"vertebrates",
"neuroscience",
"rattus",
"(rat)"
] | 2007 | A Burst-Based “Hebbian” Learning Rule at Retinogeniculate Synapses Links Retinal Waves to Activity-Dependent Refinement |
Mycobacterium tuberculosis has evolved many strategies to evade elimination by the host immune system , including the selective repression of macrophage IL-12p40 production . To identify the M . tuberculosis genes responsible for this aspect of immune evasion , we used a macrophage cell line expressing a reporter for IL-12p40 transcription to screen a transposon library of M . tuberculosis for mutants that lacked this function . This approach led to the identification of the mmaA4 gene , which encodes a methyl transferase required for introducing the distal oxygen-containing modifications of mycolic acids , as a key locus involved in the repression of IL-12p40 . Mutants in which mmaA4 ( hma ) was inactivated stimulated macrophages to produce significantly more IL-12p40 and TNF-α than wild-type M . tuberculosis and were attenuated for virulence . This attenuation was not seen in IL-12p40-deficient mice , consistent with a direct linkage between enhanced stimulation of IL-12p40 by the mutant and its reduced virulence . Treatment of macrophages with trehalose dimycolate ( TDM ) purified from the ΔmmaA4 mutant stimulated increased IL-12p40 , similar to the increase observed from ΔmmaA4 mutant-infected macrophages . In contrast , purified TDM isolated from wild-type M . tuberculosis inhibited production of IL-12p40 by macrophages . These findings strongly suggest that M . tuberculosis has evolved mmaA4-derived mycolic acids , including those incorporated into TDM to manipulate IL-12-mediated immunity and virulence .
Tuberculosis ( TB ) is the second leading cause of death from an infectious disease worldwide [1] , [2] . Mycobacterium tuberculosis is well adapted to the human host , and possesses a variety of mechanisms that promote immune evasion and thereby permit latent infection in the presence of host innate and adaptive immune responses [3] , [4] . This latent reservoir of M . tuberculosis can eventually develop into active disease when the host immune system is compromised by any of a variety of factors , the most common of which are aging , malnutrition , and concurrent infection by HIV [5]–[7] . Currently , the attenuated M . bovis strain , BCG , is the only vaccine available for routine human immunization . It has had little if any impact on the increasing global prevalence of TB , in spite of having been administered to more than a billion people [8] . Thus , work on developing new and more immunogenic vaccine candidates is crucial and requires advances in our understanding of the host-pathogen interaction . Because phagocytic cells recognize microbes before the development of specific immunity , the macrophage response to the infection is critical for the initial local containment of infection and the subsequent development of adaptive immunity . The cytokine profile produced by macrophages and other antigen-presenting cells within the first days or weeks following infection can define the type of host immunity induced , and thereby determine its effectiveness for controlling the microbial infection [9] . A critical cytokine in the control of intracellular infections is interleukin-12 ( IL-12 ) , which is produced mainly by macrophages and dendritic cells [10] . Members of the IL-12 family , including IL-12p80 , IL-12p70 , and IL-23 , are central players in various arms of early nonspecific innate immune resistance and subsequent antigen-specific adaptive immune responses to M . tuberculosis [11]–[13] . These cytokines are dimers that all share the common IL-12p40 subunit in association with a different partner , and each has a different immunoregulatory role during various stages of the immune response to intracellular pathogens [14] . For example , during innate immune response , macrophages release IL-12p80 , which is a homodimer of p40 subunits , upon initial infection to stimulate local recruitment of more macrophages . The IL-12p70 cytokine , a heterodimer of p35 and p40 subunits , induces macrophage bactericidal activity and also proliferation , cytolytic activity , and IFNγ production by NK cells during the early innate phase of the immune response . During the induction of the adaptive response , IL-12p70 produced by macrophages and dendritic cells plays a central role in polarizing T helper type 1 ( Th1 ) differentiation [10] . In addition to IL-12p70 , IL-12p80 also has a role in initiating adaptive immunity [15] . Finally , maintenance and recall responses of immunological memory require both IL-12p70 and IL-23 ( a dimer of p40 and p19 ) [16] . It is known that mice and humans with mutations in the IL-12p40 or IL-12 receptor genes are highly susceptible to mycobacterial infection , highlighting the importance of this family of cytokines in resistance to infection with these bacteria [17] , [18] . However , even in a host that is genetically normal with respect to its IL-12 axis , virulent M . tuberculosis can evade eradication . This may be explained by the fact that resistance versus susceptibility to intracellular pathogens is often determined by a delicate balance between the cytokines that initiate and those that inhibit immunity . Studies of the mechanisms of immune evasion by M . tuberculosis have revealed that one of the strategies used by the tubercle bacillus to resist eradication is to actively repress macrophage production of IL-12p40 while stimulating secretion of IL-10 , an inhibitor of IL-12 mediated-immunity [19] , [20] . This immune evasion mechanism parallels that proposed for Leishmania and Toxoplasma , another highly persistent intracellular pathogen [21] , [22] . Given previous findings on the ability of IL-12 to enhance protective immunity and extend survival in mice and humans infected with M . tuberculosis [12] , [23] , [24] , and the discovery that M . tuberculosis represses IL-12p40 production , we hypothesized that M . tuberculosis actively dampens the production of IL-12p40 cytokine in infected macrophages . To identify the mechanisms and effector molecules responsible for this , we screened a transposon library of M . tuberculosis mutants using a macrophage cell line expressing a reporter gene for monitoring IL-12p40 expression . This identified a major role in IL-12p40 repression for the mmaA4 ( methoxy mycolic acid synthase 4 ) gene , which encodes the methyl transferase that introduces oxygen-containing modifications of cell wall mycolic acids . Mutants in which mmaA4 was inactivated induced more IL-12p40 from murine macrophages , and were attenuated for virulence in mice . The attenuation of virulence was reversed in IL-12p40-deficient mice , indicating that this attenuation depended on IL-12p40-mediated immunity . Furthermore , the abundant surface and secreted glycolipid trehalose 6 , 6′-dimycolate ( TDM ) was identified as an effector molecule for the repression of IL-12p40 production by M . tuberculosis . However , purified TDM from the ΔmmaA4 mutant , which contained mycolates that were devoid of distal oxygen-containing modifications , stimulated markedly increased production of IL-12p40 and TNF-α compared to the levels resulting from stimulation with TDM from wild type M . tuberculosis . Our data identify the role of mmaA4-dependent mycolic acid modifications in the repression of IL-12p40 production , thus establishing part of the genetic and mechanistic basis for an important aspect of the immune evasion strategy of M . tuberculosis .
To screen for mutants of M . tuberculosis that were defective in repression of IL-12p40 production , we generated a macrophage reporter cell line to monitor IL-12p40 expression . Previously , we described a Raw 264 . 7 murine macrophage cell line containing a stably integrated construct of the minimal IL-12p40 promoter fused to GFP [25] . Since the cis elements required for regulation of the IL-12p40 promoter in response to M . tuberculosis infection are not known , we engineered another Raw 264 . 7 line containing a stable integration of the full-length IL-12p40 promoter fused to GFP . Flow cytometry analysis showed that GFP was not transcribed at baseline in this macrophage line , but was induced upon treatment with lipopolysaccharide ( LPS ) or infection with E . coli ( data not shown ) . Following with mycobacterial strains and species that varied in their virulence , we observed by flow cytometry the levels of GFP induction that followed a pattern similar to what was observed using a capture ELISA to quantitate IL-12 p40 levels in supernatants of similarly infected bone marrow-derived macrophages ( Figs . 1A and 1B ) . Both the GFP expression of the reporter macrophage cell line and supernatant levels of IL-12p40 in primary macrophage cultures confirmed that the avirulent M . smegmatis strain was a robust inducer of IL-12p40 production . In contrast , the virulent clinical ( Beijing HN878 ) and laboratory ( H37Rv ) strains of M . tuberculosis induced only minimally detectable transcription and secretion of this cytokine , consistent with previously reported results [26] , [27] . To explore the hypothesis that the low levels of IL-12p40 produced by macrophages infected with M . tuberculosis resulted from active inhibition of IL-12p40 transcription , we used the reporter cell line to screen for mutant bacilli that had lost this function . Since a 2-14-fold difference in GFP expression could be detected by this reporter cell line ( Fig . 1B and data not shown ) , the system was sensitive enough to allow for detection of incremental changes in the promoter activity . A library of transposon insertion mutants of the sequenced M . tuberculosis H37Rv strain was created using the Himar-1 transposon and arrayed as individual clones in 96-well plates . We used this transposon to generate M . tuberculosis mutants because it inserts randomly into frequently-occurring TA dinucleotides [28] . Approximately 2880 transposon mutants were infected individually into the macrophage reporter cells , and screened by using a fluorimetric assay to identify GFP expression that was greater than a baseline established by wild type H37Rv infection . A primary screen identified three mutants of interest , which were found by sequencing of the transposon insertion sites to have interruptions in open reading frames Rv0643c , Rv0409 , and Rv3435c . Among these candidate genes , Rv0643c is the most extensively characterized , and is annotated as the methoxy mycolic acid synthase 3 ( mmaA3 ) gene . Rv0409 is also known as ackA and encodes a putative acetate kinase , while Rv3435c encodes a predicted transmembrane protein of unknown function . A secondary screen using analysis of the infected macrophage reporter cells by FACS confirmed that the three mutants identified in the primary screening reproducibly stimulated enhanced GFP expression , with the clone bearing a transposon insertion in the mmaA3 gene showing the highest GFP expression ( Fig . 1C ) . In a tertiary screen for IL-12p40 production by ELISA of supernatant levels from bone marrow-derived macrophages infected with these mutants , the mmaA3 mutant again showed increased IL-12p40 production . In contrast , the ackA and Rv3435c mutants showed inconsistent results in the tertiary screening with primary macrophages ( data not shown ) , and thus were not analyzed further . To confirm and further evaluate the role of mmaA3 in modulating IL-12p40 expression , the mmaA3 gene was deleted from M . tuberculosis H37Rv by specialized transduction [29] . Since the transcriptional regulation of the mmaA4 gene immediately downstream of mmaA3 could also have been compromised by the transposon insertion in mmaA3 , we also generated and studied an M . tuberculosis strain with deletion in the mmaA4 gene . Using an ELISA to quantitate IL-12p40 in the medium of bone marrow-derived macrophage cultures infected with either the ΔmmaA3 or ΔmmaA4 mutants , we observed that infection of macrophages with the ΔmmaA3 mutant showed variable increases in IL-12p40 production ( data not shown ) . In addition , when grown on agar plates , the ΔmmaA3 mutant showed a mixture of both rough and smooth colony morphologies ( data not shown ) , indicating that phenotypic switching between these two morphologies was occurring frequently during culture and may have accounted for the variable effects on IL-12p40 induction . Because of this potentially confounding variable and the inconsistent effects on IL-12p40 production , no additional studies were pursued with the ΔmmaA3 mutant and all subsequent work focused on the ΔmmaA4 mutant . In contrast to ΔmmaA3 infected macrophages , bone marrow-derived macrophages infected with the ΔmmaA4 mutant showed a reproducible increase in both IL-12p40 and TNF-α production , compared to wild type M . tuberculosis . This phenotype was reversed by complementation using chromosomal insertion of a single copy of the wild type mmaA4 gene ( Fig . 2 ) . IL-12p40 and TNF-α production in macrophages infected with the ΔmmaA4 mutant increased over time , and similar cytokine increases were observed for bone marrow-derived macrophages from two different mouse strains ( i . e . , BALB/c in Fig . 2 , and C57BL/6 in Fig . S1 ) . It is known that dendritic cell also produce IL-12p40 following infection with M . tuberculosis [19] , [30] . To examine dendritic cell responses , bone marrow-derived dendritic cells were infected with the parental , mutant , or complemented strain . Dendritic cells infected with the ΔmmaA4 mutant produced copious amounts of IL-12p40 . This robust production of IL-12p40 was also observed for dendritic cells infected with wild type or complemented strain ( Fig . S2A ) . Thus , the ΔmmaA4 mutant selectively effects macrophage cytokine production . The ΔmmaA4 mutant maintained stable colony morphology with routine passage , showing a smooth colony morphology with ruffled edges when plated on media containing the detergent Tween-80 ( Fig . 3 ) . This was distinctly different from the rough colony morphology observed with similarly cultured wild type M . tuberculosis . This difference in colony morphology was reversed by complementation of the ΔmmaA4 mutant ( Fig . 3 ) . The growth rate of the ΔmmaA4 mutant in liquid culture was equivalent to that of wild type ( data not shown ) . Previous studies of mycolic acids synthesized by the mmaA4 mutant ( also known as hma , for hydroxyl mycolic acid synthase ) or by using the expression of the M . tuberculosis mmaA4 gene in M . smegmatis , provided strong evidence that the enzyme encoded by this locus is responsible for introducing the methyl and adjacent hydroxy groups on the distal meromycolate chain of the common precursor for methoxy- and keto-mycolic acids [31] , [32] . This was also supported by analysis of mycolic acid and major extractable lipids in the wild type , ΔmmaA4 mutant , and complemented strains in this study ( Figs . S3 and S4 ) . Additionally , our finding that growth of the ΔmmaA4 mutant in the presence of Tween-80 caused changes in colony morphology was consistent with a significant alteration in the lipid composition of the cell wall [33] , which would be expected given that mycolates are among the most abundant cell wall lipids . A small but significant quantity of mycolic acids are found noncovalently associated with cell wall glycolipids; the most abundant of these is TDM [34] . Since TDMs are released into the cytoplasm of macrophages infected with mycobacteria , this suggested that the alterations in cytokine production seen with the ΔmmaA4 mutant might be due to changes in the mycolic acids incorporated into its TDMs [35] . This possibility was studied by comparing the cytokine responses of macrophages to TDMs purified from wild type and from ΔmmaA4 mutant M . tuberculosis . Time course and dose response studies with purified TDM from either wild-type or the ΔmmaA4 mutant clearly showed that the ΔmmaA4 TDM mutant induced significantly more IL-12p40 and TNF-α than did wild type TDM ( Fig . 4A ) . IL-12p40 production was detected at 22 hr in conditioned media from macrophages that were treated with ΔmmaA4 TDM , and further increased 4-fold by 44 hr . TNF-α was detected at 22 hr , with no further increase thereafter ( Fig . 4B ) . In contrast , after 22 to 44 hr of incubation with wild type TDM , production of IL-12p40 remained constant and was substantially less then that stimulated by ΔmmaA4 TDM ( Fig . 4A ) . By comparison , IL-12p40 and TNF-α production increased with the addition of increasing amounts of ΔmmaA4 TDM to the cultures , whereas wild type TDM did not show a dose dependency over the range of TDM concentrations tested ( Fig . 4B ) . Similar differences in cytokine production by macrophages treated with trehalose monomycolate ( TMM ) from the mutant and wild type strains were observed ( Fig . S5 ) . These findings indicated that the increased macrophage cytokine production following infection with ΔmmaA4 mutant bacteria could potentially have been mediated by the release of modified TDMs . The differences in cytokine production observed for macrophages treated with TDM-containing oxygenated mycolic acids versus mutant TDMs lacking oxygenated mycolic acids suggested a possible inhibitory effect of the oxygenated mycolic acids on IL-12p40 responses . To test this , we analyzed the effects of mixing wild type TDM and ΔmmaA4 TDM on macrophage cytokine responses . This experiment revealed that combining the two TDMs led to significant inhibition of the IL-12p40 response stimulated by the mutant TDM alone ( Fig . 5A , left ) . Interestingly , this apparent inhibition of cytokine production was not observed for TNF-α production , which was actually slightly enhanced when the TDMs from wild type and mutant bacteria were combined ( Fig . 5A , right ) . To further assess this apparent repression of IL-12p40 production by wild type but not by ΔmmaA4 mutant TDMs , we examined the impact of each on IL-12p40 production in response to E . coli lipopolysaccharide ( LPS ) . This analysis showed that TDMs from both sources reduced the LPS-induced IL-12p40 response , but that wild type TDM was significantly more potent in this regard than was ΔmmaA4 TDM ( Fig . 5B , left ) . In contrast , we observed that neither type of TDM repressed the LPS-induced TNF-α production ( Fig . 5B , right ) . In fact , a trend was observed of increased stimulation of TNF-α production by the mutant , as compared to the wild type , TDM . Taken together , these data suggested that the TDM of wild type M . tuberculosis was a significant mediator of IL-12p40 repression in macrophages , whereas the TDMs of the ΔmmaA4 mutant , which lack keto- and methoxy-mycolates , were much less potent in this regard . In addition , it was apparent from these results that wild type TDMs possessed both weak stimulatory and strong inhibitory activity for macrophage production of IL-12p40 , whereas the ΔmmaA4 TDM was significantly more stimulatory and had reduced capacity to inhibit IL-12p40 responses . The increased IL-12p40 and TNF-α observed in macrophages infected with ΔmmaA4 M . tuberculosis or treated with its TDM could be hypothesized as potentially having either of two opposing outcomes in vivo . The increases in these cytokines could either induce a protective immune response and improved host survival , or could lead to a deleterious exacerbation of immunity with tissue damage , contributing to poor outcome . To distinguish between these two possibilities , we infected immunocompetent C57BL/6 mice with the ΔmmaA4 mutant and compared their survival rate to that of mice infected with wild type ( H37Rv ) M . tuberculosis or the complemented ΔmmaA4 strain . As shown in Fig . 6A , all mice infected with wild type M . tuberculosis died by approximately 225 days post-infection . In contrast , all mice infected with the ΔmmaA4 mutant survived 450 days post-challenge , while complementation of ΔmmaA4 mutation restored virulence to the mutant . To examine whether the attenuation in virulence observed for mice infected with the ΔmmaA4 mutant was dependent on IL-12p40 , we assessed the survival rate of IL-12p40 deficient mice infected with wild type M . tuberculosis , the ΔmmaA4 mutant , or the complemented mutant . IL-12p40-deficient mice infected with wild type M . tuberculosis or the complemented mutant strain all died between 45 and 60 days . Noticeably , IL-12p40-deficient mice infected with the ΔmmaA4 mutant survived only slightly longer ( p = 0 . 08 by Logrank test , compared to wild type M . tuberculosis infected animals ) , with all of these animals dying by 62 days post-infection ( Fig . 6B ) . This finding was consistent with the conclusion that the attenuation of the ΔmmaA4 mutant in vivo was dependent on the presence of a normally functioning IL-12p40 gene as well as on the ability of this mutant to elicit a more robust IL-12 response . This in vivo study provided additional support for the view that the repression of IL-12p40 by mycolic acids with oxygen-containing modifications plays a major role in immune evasion that leads to the virulence of M . tuberculosis .
M . tuberculosis has evolved strategies to evade the antimicrobial effects of IL-12-induced immunity , including selective repression of IL-12p40 expression in macrophages [19] , [20] . To identify the factor ( s ) involved in this evasion strategy , we screened for mutants that induced IL-12p40 , and identified the M . tuberculosis mmaA4 gene as a key locus involved in modulation of IL-12p40 . We demonstrated that infection of macrophages with the ΔmmaA4 mutant of M . tuberculosis H37Rv resulted in production of more IL-12p40 . In addition , our results strongly suggest that this enhanced induction of IL-12p40 could be mediated by the mycolate-containing glycolipid TDM , which is known to be secreted as a potential immunomodulator into the cytosol of infected macrophages . Consistent with this view , we also showed that TDM from wild type M . tuberculosis repressed macrophage IL-12p40 production . To our knowledge , this is the first demonstration of such inhibitory activity for TDMs of M . tuberculosis , and also the first data to implicate the methoxy- and keto- modifications of the mycolates in TDM in the expression of this inhibitory activity on macrophages . Consistent with findings reported in the literature that M . tuberculosis does not repress IL-12 production in dendritic cells , we found that dendritic cells produced similar amounts of IL-12 , whether infected with the wild type , the complemented strain , or the ΔmmaA4 mutant . Additionally , dendritic cells treated with TDM produced levels of IL-12 comparable to those resulting from LPS treatment . Thus , the mmaA4 mutation has a selective effect on macrophages . It is interesting to note that suppression of IL-12 production in macrophage , but not dendritic , cells as an immune evasion mechanism has also been observed for other well-characterized persistent intracellular pathogens , such as , Leishmania and Toxoplasma [21] , [22] . Two independent reports have shown that constitutive expression of the IL-12p40 gene in mice did not improve host immunity against M . tuberculosis [36] , [37] . Nevertheless , IL-12 is necessary and sufficient for achieving normal levels of protective host immunity against M . tuberculosis and other mycobacteria [11] , [12] , [24] . This disparity between a protective response when IL-12 is produced at the time of infection and a lack of enhanced immunity when IL-12 is constitutively expressed , underscores the importance of temporal and spatial regulation of IL-12p40 expression during infection . Thus , any perturbation of such regulation , such as constitutive and generalized expression , or M . tuberculosis-mediated repression , could shift the balance toward a suboptimal induction of host immunity . In the current study , we showed that the ability of M . tuberculosis to interfere with normal production of IL-12p40 was dependent on its production of mycolic acids with distal chain keto or methoxy groups , and that TDMs containing such modified mycolates could be the effector molecules for this immune evasion mechanism of M . tuberculosis . Specifically , our data suggest that M . tuberculosis may have evolved keto- and methoxy-modification of the mycolic acids incorporated into TDM in order to manipulate IL-12p40-mediated immunity in the host macrophages . We predicted that the removal of such IL-12p40 inhibitory components of M . tuberculosis would lead to a decrease in bacterial burden and would increase host survival . Indeed , Dubnau et al . previously showed attenuation of growth of the ΔmmaA4 mutant ( also known as hma ) in a mouse model of M . tuberculosis infection . Extending this observation , our data demonstrated that mice infected with the ΔmmaA4 mutant survived 400 days , at which time the experiment was terminated . Moreover , the attenuation depended on IL-12p40-mediated immunity . As part of the mechanism of M . tuberculosis pathogenesis , our results provide new insights into the link between oxygenated mycolic acids on TDM and the suppression of IL-12p40-mediated immunity contributed by macrophages . Mycolic acids are long-chain α-alky ß-hydroxy fatty acids unique to mycobacteria , and they comprise approximately 30% of the dry weight of M . tuberculosis [34] . Structurally , the mycolic acids can be broadly distinguished into two classes , de-oxygenated ( cyclopropyl ) and oxygenated ( keto and methoxy ) , based on the chemical functional group at the distal position of their long meromycolate chains . Several cyclopropane synthetases and methyl transferases are involved in the introduction of these functional groups , including the so-called methoxy mycolic acid synthases encoded by the mmaA3 and mmaA4 genes that were the focus of our current study . While mycolates are found covalently attached to the arabinogalactan of the M . tuberculosis cell wall , they also have an important role as a component of several extractable cell wall glycolipids , such as TMM and TDM . Significantly , both of these glycolipids are shed from the mycobacterium into the cytoplasm of infected macrophages , and they are widely believed to play a role in modulating many of the cellular processes that occur in the M . tuberculosis-infected mammalian host [35] . Recently , there has been increased appreciation of the specific bioactivities associated with each functional group . For example , TDM purified from the ΔcmaA2 mutant of M . tuberculosis which lacks trans-cyclopropanation of mycolic acids stimulated increased TNF-α production by macrophages [38] . On the other hand , TDM which lacks both cis- and trans-cyclopropanation , as do those isolated from the ΔpcaA mutant , caused delayed TNF-α production [39] . Interestingly , cyclopropanation of mycolic acid affects only TNF-α , but not IL-12p40 , production by macrophages . Our data extend earlier characterizations of the biological activities of TDM and explore the previously uncharacterized role of the distal keto and methoxy groups that are missing from the mycolic acids incorporated in TDMs of the ΔmmaA4 mutant . Previously , Oswald et al . reported that peritoneal macrophages isolated from mice showed activation of transcription of IL-12p40 ex vivo when treated with TDM purified from M . tuberculosis [40] . The findings of these investigators are consistent with our observation that wild type TDM induced low levels of IL-12p40 production by macrophages . In addition , by comparing the effects of wild type and ΔmmaA4 TDMs on macrophage production of IL-12p40 , we uncovered an underappreciated novel repressing function for wild type TDM . This was evident in the ability of wild type TDM to significantly dampen LPS-induced IL-12p40 production . Since LPS signals mainly through the TLR4 receptor and wild type TDM bioactivity has been found to be independent of TLR4 signaling , it is unlikely that inhibition of LPS-mediated IL-12p40 production resulted from competition between wild type TDM and LPS for the TLR4 receptor [41] . In comparison , ΔmmaA4 TDM was attenuated in its ability to repress induction of IL-12p40 by LPS , and ΔmmaA4 TDM was also more directly stimulatory than wild type TDM with regard to activating macrophage production of IL-12p40 . It is likely that the different biological activities observed for TDMs from wild type versus those from ΔmmaA4 bacteria were based on the chemical and structural differences conferred by the functional groups of their mycolates . In particular , while the lack of methoxy- and keto- groups may have been responsible for the loss of ability to repress the activation of IL-12 transcription , it is also possible that the novel appearance of epoxy mycolates that we observed in the ΔmmaA4 mutant TDM accounted for at least part of its increased stimulatory effect . Although entirely speculative at this point , this possibility is suggested by the fact that the epoxide functional group is highly reactive , and may thus interact more avidly with cellular components that are normally not engaged during M . tuberculosis infection . At present , it is not technically feasible to purify TDMs based on their precise mycolate composition into separate homogeneous groups , and future studies using chemically synthesized TDMs with precisely fixed mycolate structures may be required to clarify the cytokine-inducing or -repressing activity of each mycolic acid functional group when associated with different TDMs . Previous studies focused on transcriptional profiling of M . tuberculosis in the lungs of infected mice have shown that the mmaA4 gene is upregulated in vivo , compared to its level in bacteria growing in culture [42] . This observation suggests the interesting possibility that M . tuberculosis remodels its mycolic acid composition as a counter-response to host immunity . In support of this idea , analysis of mycolic acid production during infection has shown that M . tuberculosis synthesizes more keto-mycolates following macrophage infection [43] . Evading host immunity by modifying bacterial components that interact with the host is a strategy common among opportunistic bacteria that cause chronic infection . For example , Pseudomonas aeruginosa ( associated with cystic fibrosis ) and Porphyromonas gingivitis ( associated with periodontal disease ) have naturally occurring variants of LPS structures that antagonize cytokine production [44] , [45] . Additionally , Helicobacter pylori ( associated with peptic ulcer disease ) flagellin contains natural modifications which allow the bacterium to evade detection by the immune system [46] . A prominent theme emerging from studies of immunologically active glycolipids in M . tuberculosis is that this pathogen has evolved a number of mechanisms for modifying these compounds to reduce their recognition by the innate immune system and dampen their tendency to stimulate cytokine production . In addition to the modification of mycolic acids in TDM , the modification of M . tuberculosis lipomannan ( LM ) by attachment of a large arabinan to generate lipoarabinomannan ( LAM ) also may represent a strategy designed to block the ability of a mycobacterial glycolipid to activate IL-12p40 production [25] , [47] . Similarly , the phenol group on phenolic glycolipid alters cytokine production of macrophages . Absence of this phenol group in mutants lacking the pks 1–15 gene cluster abrogates cytokine-repressing activity , and leads to attenuation of virulence with extended survival in mouse infection studies [48] . Our current findings add to a growing literature demonstrating that M . tuberculosis has evolved a repertoire of molecules that disrupt macrophage effector mechanisms . In addition to our current findings for TDM , the ESAT-6 protein of M . tuberculosis also suppresses IL-12p40 [49] , [50] . This suppression of macrophage IL-12 production is likely to be responsible for increasing bacterial survival during innate immune response , given the critical role of IL-12p40 as a macrophage chemoattractant and in interferon gamma production [51]–[54] . By expressing various IL-12 inhibitors that might function at different times and locales during the course of the infection , M . tuberculosis is well adapted to survive even in the face of a normal host immune response . We believed that additional bacterial components of M . tuberculosis involved in IL-12p40 repression could be revealed by extending our screening approach to saturation . The identification and removal of mycobacterial genes involved in the inhibition of important cytokine responses , such as we have demonstrated in the current study for mmaA4 , should provide a straightforward and rational approach for creating more immunogenic strains of attenuated mycobacteria that may ultimately yield more effective vaccines and immunotherapies for the prevention of tuberculosis .
Cultures of mycobacteria were routinely grown in 7H9-C media which contained Middlebrook 7H9 media supplemented with OADC ( oleic acid/albumin/dextrose/catalase ) ( Difco , Becton-Dickinson ) , 0 . 5% Glycerol , and 0 . 05% Tween-80 . Colony morphologies for wild type and mutant M . tuberculosis strains were observed by plating bacterial cultures on Middlebrook 7H10 plates supplemented with OADC , 0 . 5% glycerol , and 0 . 05% Tween-80 . For mutant strains , 50 µg/ml of hygromycin was included in the media . Six-to-eight-week-old female BALB/c mice were purchased from Jackson Laboratory ( Bar Harbor , ME ) . Bone marrow cells were flushed with phosphate buffered saline ( PBS ) from the femurs of mice and cultured in Dulbecco's Modified Eagle Medium ( DMEM ) , supplemented with 10% heat-inactivated fetal calf serum ( FCS ) plus 20% conditioned medium from a culture of L929 cells ( as a source of M-CSF ) , for 7 days at 37°C , 5% CO2 . Bone marrow-derived macrophages were harvested on day 6 to plate for infection with different M . tuberculosis strains on day 7 . Bone marrow-derived dendritic cells were differentiated by methods described by Lutz , MB et al . Briefly , bone marrow cells were seeded at 3×105 cells/100 mm petri dish . The cells were differentiated in the presence of 10 ng/ml GM-CSF ( Peprotech ) . Media were changed every two days . Cells ( 1 . 5×105/200 ul media ) were seeded into 48 wells on day 7 and infected on day 8 . The construction of the −350+55 IL-12p40-GFP Raw 264 . 7 reporter line was described previously [25] . Using a similar strategy to monitor IL-12p40 expression from a full-length ( FL ) IL-12p40 promoter [55] , position −800 to +55 relative to the transcription start site of the IL-12p40 promoter was amplified from C57BL/6 mouse genomic DNA by PCR by using upstream primer 5′ACAGGATTGCACACCTCTTTG 3′ and downstream primer 5′ TTGCTTTGCTGCGAGC3′ . The 856 bp PCR product was placed into the TOPO cloning vector ( Invitrogen ) to create the plasmid pFL . IL-12p40 . TOPO . The full-length IL-12p40 GFP reporter construct ( pFL . IL-12p40 . EGFP-1 ) was generated by ligating the HindIII/PstI fragment from the pFL . IL-12p40 . TOPO into the HindIII and PstI cleaved enhanced green fluorescent protein reporter vector , pEGFP-1 ( BD Bioscience ) . Raw 264 . 7 cells were stably transfected with pFL . IL-12p40 . EGFP-1 using electroporation , as described previously [25] . Following selection with G418 ( 1 mg/ml ) , a stable −800+55 IL-12p40-GFP Raw 264 . 7 macrophage cell line was cloned by limiting dilution under G418 selection and maintained in DMEM with high glucose , supplemented with 10% FCS , 10 mM HEPES , and 1 mg/ml of G418 . Clones that express GFP only when treated with LPS ( 10 ng/ml ) or CpG ( 100uM ) were expanded for use . For each infection , a new vial of bacterial culture was thawed from stocks kept at −70°C . Thawed M . tuberculosis H37Rv , Beijing HN878 , or M . smegmatis were grown in 10 ml 7H9-C medium , as described above . The ΔmmaA4 mutant was grown in 7H9-C medium along with 50 µg/ml hygromycin , while the complemented ΔmmaA4 mutant was grown in 7H9-C medium with 40 µg/ml of kanamycin . All mycobacterial strains were grown to an OD600 nm of between 0 . 1 and 0 . 3 prior to infection , since a population of the mycobacterial culture will autolyse when grown to OD ≥0 . 5 undergo autolysis ( data not shown ) [56] . Prior to infection , the bacteria were pelleted and resuspended in 7H9-C medium . The resuspended pellets were treated once with 10 sec of continuous sonication to minimize aggregation . Male and female C57BL/6 and IL-12p40−/− mice , 6 to 10 weeks of age , were acquired from Jackson Laboratories . One ml aliquots of frozen suspensions of M . tuberculosis , H37Rv , ΔmmaA4 mutant , or complemented ΔmmaA4 mutant were thawed and innoculated into 7H9-C media containing the appropriate selecting agents ( 50 µg/ml hygromycin for ΔmmaA4 , and 40 µg/ml kanamycin for the complemented mutant ) . Bacteria from frozen stocks were grown to an OD600nm of between 0 . 1 and 0 . 3 , and then collected by centrifugation and washed once with PBS- 0 . 05% Tween-80 . Cell pellets were resuspended to 1×107 CFU/ml , and 20 µl of 1∶5 diluted Antifoam ( Sigma ) was added to 10 ml of the bacteria suspension to prevent froth formation during aerosalization . The bacterial suspension was placed in the nebulizer jar of a whole-body exposure aerosol chamber ( Mechanical Engineering Workshop , Madison , WI ) . Mice were exposed for 20 min , with a chamber purge time of 30 min between strains . 24 hr post-aerosalization , lungs from 3 mice per group were harvested to determine the inoculum per group . Bone marrow-derived macrophages or dendritic cells were seeded in triplicate at 2×105 per well ( for macrophages ) or 1 . 5×105 per well ( for dendritic cells ) in 48-well plates , or 2×105 per well in 96-well plates for the FL . IL-12p40-GFP macrophage reporter cell line . The macrophages were infected with mycobacteria at a multiplicity of infection ( MOI ) of 3 or 10 . After 4 hr incubation in a humidified incubator at 37°C in the presence of 5% CO2 , non-ingested bacteria were removed by washing gently 3 times with pre-warmed DMEM-C medium for macrophages and with RPMI-C medium for dendritic cells . Each well then received 200 µl DMEM-C or RPMI-C containing 50 µg/ml gentamicin ( to kill the remaining extracellular bacteria ) , and plates were cultured in a humidified incubator at 37°C in the presence of 5% CO2 . Infection was allowed to proceed for 16 to 24 hr before cell supernatants were harvested . For time course studies , the supernatants were collected at the additional time points of 48 and 72 hr . Supernatant was filtered with 0 . 22 µm SpinX columns ( Costar ) to remove any uningested extracellular bacteria . Cytokines in the conditioned medium were analyzed by sandwich ELISA using the Biosource International ( Camarillo , CA ) kit for IL-12p40 and TNF-α , following the manufacturer's protocol . Following infection of the FL . IL-12p40 GFP- reporter macrophage cell line with H37Rv , HN878 Beijing , and M . smegmatis , the experiment was allowed to proceed for 16 to 24 hr before processing the cells for FACS analysis . Mycobacteria infected cells were trypsinized , fixed with equal volume of 4% paraformaldehyde , and left at 4°C overnight . The following day , GFP expression was ascertained by using the FACSCalibur flow cytometer ( BD Biosciences ) and analyzed with FlowJo software ( Tree Star ) . The Himar-1 M . tuberculosis H37Rv mutant library was generated using the Himar-1 transposon delivered by phage , pHAE159 , as described previously [28] , [29] . Briefly , the phage-containing mariner transposon was propagated to high titer in MP buffer ( 50 mM Tris ( pH 7 . 6 ) , 150 mM NaCl , 10 mM MgCl2 , 2 mM CaC12 ) and used to transduce the M . tuberculosis H37Rv strain . The transductions were plated on 7H10 plates containing 50 µg/ml hygromycin , and placed at 37°C for three weeks . Transductants were picked into 96-well plates containing 200 µl of 7H9 media supplemented with OADC , 0 . 5% glycerol , 0 . 05 % Tween-80 , and 50 µg/ml hygromycin . A Himar-1 transposon library of M . tuberculosis H37Rv was grown to late-log phase . Aliquots of the mariner M . tuberculosis H37Rv library were made into separate 96-well plates for stocking and were diluted and grown to mid-log phase for screening . Individual clones from the M . tuberculosis mariner transposon library were grown in wells of 96-well plates with monitoring of cell density by photometric measurements of optical density ( OD ) at 590 nm using a 96-well plate reader ( Victor II plate reader , Perkin Elmer ) . After 2 days of growth , each well was diluted to approximately 2×106 CFU per 10 µl . The IL-12 reporter strain , Raw 264 . 7- FL . IL-12p40-GFP , was seeded at 2×105 per well in 96-well plates the day before infection with clones from the mariner transposon library . An aliquot of 10 µl of bacteria from each well was used to infect the FL . IL-12p40-GFP Raw 264 macrophage reporter cell line ( i . e . , an MOI of 10 ) . After incubation for 4 hr in a 5% CO2 humidified incubator at 37°C , non-ingested bacteria were removed by washing gently 3 times with pre-warmed DMEM-C medium . Each well then received 200 µl DMEM-C containing 50 µg/ml gentamicin , and the plates were cultured as before for an additional 16 hr at 37°C , at which time IL-12 expression was found to be maximal . The GFP expression from individual wells on the plate was determined by the use of the Viktor II plate reader using a 488 nm/530 nm excitation/emission filter pair and reading for 1 . 0 sec per well . For secondary screening of the candidates , the mutants were expanded in 10 ml cultures , and grown to an OD600nm of between 0 . 1 and 0 . 3 . The IL-12 reporter macrophages were infected with each clone in duplicate and incubated overnight , as described above . After 16 hr , the cells were harvested by trypsinization , and single-cell suspensions from these infected macrophages were generated . An equal volume of 4% paraformaldehyde was added to each well to allow fixation overnight at 4°C , and flow cytometry analysis for GFP expression was performed the following day . Standard genomic DNA ( gDNA ) preparations were made for transposon insertion mutants . Briefly , 10 ml cultures were grown to an OD600nm of between 0 . 5 and 0 . 7 , and then centrifuged and the pellets extracted for gDNA . Ten µl aliquots of gDNA were digested with BssHII in 50 µl for 1 hour , after which 4 µl aliquots of digested gDNA were self-ligated for 1 hr using the Rapid Ligation Mixture kit from Roche Laboratories . Five µl of ligation mixture was transformed into competent DH5α pir bacteria and selected for on LB plates containing 150 µg/ml hygromycin . The exact location of each transposon insertion site in the selected mutants was determined by sequencing the flanking M . tuberculosis gDNA: Upstream flanking sequence ( 5′ -AGAATAGACCGAGATAGGGT ) , Downstream flanking sequence ( 5′-ACTTTAGATTGATTTCGCGT ) . The mmaA3 ( Rv0643c ) and mmaA4 ( Rv0642c ) mutants were constructed by homologous recombination using specialized transducing phages [29] . The deletion phagemid for the ΔmmaA3 mutant was constructed by PCR amplification of the 5′-flanking region of mmaA3 using M . tuberculosis H37Rv genomic DNA with the following primer pairs: 0643cRL 5′ TTTTTTTTCCATAGATTGGTCACTCGATCACCGGCTTGCACGTA 3′ and 0643cRR 5′TTTTTTTTCCATCTTTTGGGGAGACGTCGTAGTGCGCTTGGATG 3′ . This PCR product was 553 bp . For the 3′ flanking region of mmaA3 , the following primer pairs were used: 0643c LL 5′TTTTTTTACCATAAATTGGGGAACAGTCGGCGAAGACGGGTTT 3′ and 0643cLR 5′ TTTTTTTTCCATTTCTTGGTGAAGTTGGCCCAGTCGCTCAGCAG 3′ . This PCR product was 811 bp . The deletion phagemid for the ΔmmaA4 mutant was constructed by PCR amplification of the 5′-flanking region of mmaA4 from M . tuberculosis H37Rv genomic DNA using the primer pairs 0642cRL 5′ TTTTTTTTCCATAGATTGGTTCGAGACGGCGCGTTTCATCA 3′ and 0642cRR 5′ TTTTTTTTCCATCTTTTGGCGACCCGCGTAAGGCAGACCAG 3′ for the 5 prime arm . This PCR product was 994 bp . The primer pairs were 0642cLL 5′TTTTTTTACCATAAATTGGAGCACTCGATCACCGGCTTGCACGTA3′ and 0642cLR 5′TTTTTTTTCCATTTCATGGTCCAACCGCACCCAATGTCCAGCAG 3′ for the downstream arm , which gave rise to a 723 bp PCR product . Following cloning into p0004S ( 0642c . p004S or 0643c . p004S ) , the resulting plasmid was then packaged into the temperature-sensitive phage phAE159 , as described earlier , to yield the knockout phages for mmaA3 ( phAE301 ) and mmaA4 ( phAE302 ) . Specialized transduction was performed , as described previously [29] , and the transduction mix was spread on 7H10 plates , selecting with 50 µg/ml hygromycin . Hygromycin-resistant clones were screened for deletion by Southern analysis . Briefly , gDNA from mmaA3 or mmaA4 mutants was digested with StuI . Deletion analysis for the ΔmmaA3 mutant was confirmed by probing the southern blot with the PCR product ( 811 bp ) from the primer pairs 0643c LL & 0643cLR . Following homologous recombination , the mmaA3 mutant had a 2742 bp fragment as compared to wild type , which gave a 6127 bp fragment . Deletion analysis for the ΔmmaA4 mutant was confirmed by probing the southern blot with the PCR product ( 723 bp ) from the primer pairs 0642cLL & 0642cRL . Following homologous recombination , the ΔmmaA4 mutant had a 3542 bp fragment as compared to wild type , which gave a 6127 bp fragment . Complementation analyses were performed with the cosmid 3E2 ( Rv0630c–Rv0654c ) , which contained the mmaA4 gene in the integration-proficient vector pYUB412 . The transformation of the mutant strains with the constructs by means of electroporation was described previously . Kanamycin-resistant clones were screened for reversion of mutant colonial morphology . Initially , 10 ml cultures of wild type M . tuberculosis , ΔmmaA4 mutant , or complemented ΔmmaA4 mutant at an OD 600 nm∼0 . 4 were labeled using 1 µCi/ml [14C]-acetic acid and further incubated for 12 hr . Cells were recovered by centrifugation at 27 , 000×g for 10 min and carefully freeze-dried using a Savant SpeedVac . Cellular-associated lipids were extracted twice using 2 ml of CHCl3/CH3OH/H2O ( 10:10:3 , v/v/v ) for 3 hr at 50°C . Organic extracts were combined with 1 . 75 ml CHCl3 and 0 . 75 ml H2O , mixed and centrifuged . The lower organic phase was recovered , washed twice with 2 ml of CHCl3/CH3OH/H2O ( 3:47:48 , v/v/v ) , and then dried and resuspended with 200 µl of CHCl3/CH3OH ( 2:1 , v/v ) . The residual cell pellet was subjected to alkaline hydrolysis using 15% aqueous tetrabutylammonium hydroxide ( TBAH ) at 100°C overnight , followed by the addition of 4 ml of dichloromethane , 300 ml iodomethane , and 4 ml of water . The entire reaction mixture was then mixed for 1 hr . The upper aqueous phase was discarded and the lower organic phase washed twice with water and evaporated to dryness . Mycolic acid methyl esters ( MAMES ) were re-dissolved in diethyl ether . After centrifugation , the clear supernatant was again dried and resuspended in dichloromethane ( 100 ml ) and an aliquot subjected to 1-dimensional High Performance Thin-Layer Chromatography ( 1D-HPTLC ) , using two developments of hexane/ethyl acetate [95:5] ) . MAMES were visualized by autoradiography by exposure of TLCs to X-ray film ( Kodak X-Omat ) . Four liter cultures of wild type M . tuberculosis or ΔmmaA4 mutant were grown to OD600nm = 0 . 4 . Mycobacteria were recovered by centrifugation at 3000 RPM for 15 min in a table-top centrifuge . Cellular lipids were extracted twice , as described above , from freeze-dried cells using 200 ml of CHCl3/CH3OH/H2O ( 10:10:3 , v/v/v ) for 3 hr at 50°C . Organic extracts were combined with 175 ml CHCl3 and 75 ml H2O , mixed and centrifuged . The lower organic phase was recovered , washed twice with 200 ml of CHCl3/CH3OH/H2O ( 3:47:48 , v/v/v ) , dried , and resuspended with 2 ml of CHCl3/CH3OH ( 2:1 , v/v ) . The lipid extract was examined by 2-dimensional TLC on aluminum-backed plates of silica gel 60 F254 ( Merck 5554 ) , using chloroform/methanol/water ( 100:14:0 . 8 , v/v/v ) in the first direction and chloroform/acetone/methanol/water ( 50:60:2 . 5:3 , v/v/v ) in the second direction . TDM and TMM were visualized either by spraying plates with α-naphthol/sulfuric acid , or by spraying with 5 % ethanolic molybdophosphoric acid , followed by gentle charring . The crude lipid extract ( 250 mg ) dissolved in chloroform/methanol ( 2:1 , v/v ) was applied to a diethylaminoethyl ( DEAE ) cellulose column ( 2 cm×15 cm ) and the flow-through kept for further purification . TDM and TMM were further purified by preparative TLC on 10 cm×20 cm plastic-backed TLC plates of silica gel 60 F254 ( Merck 5735 , Darmstadt , Germany ) , run in chloroform/methanol/ammonium hydroxide ( 80:20:2 , v/v/v ) . The plates were then sprayed with 0 . 01% 1 , 6-diphenyl-1 , 3 , 5-hexatriene dissolved in petroleum ether/acetone ( 9:1 , v/v ) , and lipids were visualized under UV light . Following detection , the plates were re-developed in toluene to remove diphenyl-1 , 3 , 5-hexatriene , and the corresponding TDM and TMM bands were scraped from the plates and extracted from the silica gel using 3 extractions of chloroform/methanol ( 2:1 , v/v ) to provide highly purified TDM and TMM . Quantitation of the purified TDM and TMM was done by directly weighing the material . The highly purified TDM and TMM from wild type M . tuberculosis was reconstituted in petroleum at a concentration of 200 µg/ml . Aliquots of 500 µl were dispensed into endotoxin-free glass vials , and the samples were dried under nitrogen for storage . The TDM and TMM stock was tested for endotoxin contamination using the Limulus Amoebocyte Lysate ( LAL ) assay from Bio Whittaker , following the manufacturer's protocol . Briefly , the TDM ( or TMM ) in one of the vials was resuspended in DMSO to a Cf = 1 mg/ml . Ten µl of the sample was used in the LAL assay . The TDMs and TMMs from wild type M . tuberculosis H37Rv or ΔmmaA4 mutant were endotoxin-free ( data not shown ) . At the time of the experiment , 100 µg of TDM was reconstituted in 500 µl of petroleum ether . A series of 2-fold dilutions of TDM was made with petroleum ether to yield 10 µg/100 µl , 5 µg/100 µl , and 2 . 5 ug/100 µl , after which 100 µl of each dilution was used to coat a 48-well plate . The plate was air dried to evaporate the solvent , washed once with PBS , and then air dried again . The TDM dose used in this assay was higher than that typically used for pathogen glycolipids from gram-negative bacteria such as LPS , but comparable to that used for glycolipids and other cell wall-associated immune activators of gram-positive bacteria , such as lipoteichoic acid and peptidoglycan [57]–[59] . Bone marrow-derived macrophages were then immediately added at 2×105 cells/200 µl per well in a 48-well plate . For wild type TDM and ΔmmaA4 TDM cotreatment , 10 µg of wild type TDM and 5 µg of mmaA4 TDM were added in 100 µl each of petroleum ether to the same wells of a 48-well plate . The contents were mixed to ensure even distribution of the lipids before the plate was air-dried , washed with PBS , and then air dried again , before the addition of macrophages . For E . coli LPS and wild type TDM cotreatment , the wells were first coated with 5 µg wild type TDM , air dried , washed with PBS , and then air dried again . This was followed by the addition of bone marrow-derived macrophages and , 16 hr later , 100 ng/ml E . coli LPS . Culture supernatants were harvested , filtered , and then analyzed by ELISA for cytokine levels , as described above . TDM from M . tuberculosis purchased from Sigma was also tested . The IL-12p40 response of macrophage and dendritic cells is similar to that of TDM purified by us from wild type M . tuberculosis . TDM studies reported in Fig . S2 was purchased from Sigma . Supplementary materials and methods for total lipid extraction and analysis can be found in Protocol S1 . mmaA4 METHOXY MYCOLIC ACID SYNTHASE 4 ( HYDROXY MYCOLIC ACID SYNTHASE ) [Mycobacterium tuberculosis H37Rv] GeneID: 888056 mmaA3 METHOXY MYCOLIC ACID SYNTHASE 3 [Mycobacterium tuberculosis H37Rv] GeneID: 1091772 ackA ACETATE/PROPIONATE KINASE [Mycobacterium tuberculosis H37Rv] GeneID: 886399 Rv3435c PROBABLE CONSERVED TRANSMEMBRANE PROTEIN [Mycobacterium tuberculosis H37Rv] GeneID: 887564 | Currently , one-third of the world's population has tuberculosis ( TB ) . TB , an ancient foe , has reemerged to become a threat to global public health . A central problem in TB research is to investigate why the host immune system cannot sterilize the infection caused by the bacterium Mycobacterium tuberculosis . Interleukin-12 ( IL-12 ) , a molecule produced by macrophages in response to pathogens , plays an important role in orchestrating sterilizing immunity . However , M . tuberculosis has evolved mechanisms that block IL-12 production and thereby assist the bacterium in establishing chronic infection . We discovered that mutation of the mycobacterial mmaA4 gene , which controls the chemical modification of complex lipids of M . tuberculosis called mycolic acids , renders the bacterium unable to block IL-12 production . Mycolic acids incorporated into a secreted bacterial molecule called trehalose dimycolate ( TDM ) from M . tuberculosis had the ability on their own to suppress the production of IL-12 by activated macrophages; we also showed that TDM from the mmaA4 mutant of M . tuberculosis is attenuated for suppression . Our results identify a critical part of the genetic basis and mechanism for an important immune evasion function in M . tuberculosis , and should contribute to the design of future vaccines and immunotherapies for disease caused by this pathogen . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics/gene",
"discovery",
"immunology",
"microbiology",
"infectious",
"diseases"
] | 2008 | Mycolic Acid Modification by the mmaA4 Gene of M. tuberculosis Modulates IL-12 Production |
Although information regarding morphogenesis of the hepatitis C virus ( HCV ) is accumulating , the mechanism ( s ) by which the HCV genome encapsidated remains unknown . In the present study , in cell cultures producing HCV , the molecular ratios of 3’ end- to 5’ end-regions of the viral RNA population in the culture medium were markedly higher than those in the cells , and the ratio was highest in the virion-rich fraction . The interaction of the 3’ untranslated region ( UTR ) with Core in vitro was stronger than that of the interaction of other stable RNA structure elements across the HCV genome . A foreign gene flanked by the 3’ UTR was encapsidated by supplying both viral NS3-NS5B proteins and Core-NS2 in trans . Mutations within the conserved stem-loops of the 3’ UTR were observed to dramatically diminish packaging efficiency , suggesting that the conserved apical motifs of the 3´ X region are important for HCV genome packaging . This study provides evidence of selective packaging of the HCV genome into viral particles and identified that the 3’ UTR acts as a cis-acting element for encapsidation .
It is known that positive strand RNA viruses package their genome via replication-coupled processes [1–3] . Nevertheless , for several viruses , viral particle assembly occurs via recognition of particular sequences or structures termed packaging signals that are unique to the viral genes . These signals are more conserved than other parts of the viral genomes and are usually highly structured . However , thus far , a packaging signal has not been identified for any member of the Flaviviridae family . The genome of HCV is a positive strand RNA with a single open reading frame flanked by highly conserved and structured untranslated regions ( UTRs ) at each end [4] . The 5’ UTR contains determinants for cap-independent translation and cis-acting elements for RNA replication . The 3’ UTR contains a short variable region , a poly ( U/UC ) tract with an average length of 80 nt ( nucleotide ) and a conserved 98-nt X-tail region ( 3’ X ) . Direct interactions of the 3’ UTR and the translation machinery facilitate efficient initiation of subsequent translation [5] . Mutations in the 3’ X region were shown to abort replication [6 , 7] , which illustrates the important role of the 3’ UTR in replication . Both the length and the composition of the poly ( U/UC ) tract are critical for HCV genome replication [8] . In this study , by using replication-defective trans-packaging systems , we identified that the 3’ UTR of the HCV genome acts as a cis-acting element for RNA packaging . Within the 3’ UTR , the loop sequences of stem-loop structures appear to be essential for HCV RNA packaging .
Excess amounts of the 5’ end of subgenomic HCV RNA species , which presumably resulted from premature termination of transcription or RNA cleavage in the viral genome , have been detected in the livers and sera of hepatitis C patients [9 , 10] . To elucidate the properties of HCV RNAs throughout replication and encapsidation , we investigated the distribution of viral RNAs in a HCV cell culture ( HCVcc ) system . We hypothesized that , if certain viral RNA species are selectively incorporated into virions , then different 3’- to 5’-end molecular ratios of HCV RNA would be observed in virion-rich fractions of the culture supernatant compared to those in the whole cell supernatant and in the cells . HCV RNAs were determined using Quantitative ( q ) RT-PCR reactions targeting the 5’ UTR ( 5’ end ) and the NS5B region approaching the 3’ UTR ( 3’ end ) ( S1A Fig ) . The molecular ratios of 3’- to 5’ end were then calculated using RNA copy numbers of NS5B and the 5’ UTR . Firstly , serial dilutions of in vitro synthesized HCV RNA of approximately full-genome length , IVT-F ( S1B Fig ) , were analyzed using these two qRT-PCRs . RNA copy numbers of each dilution were determined by targeting the 5’ UTR and the NS5B region ( S1C Fig ) . The ratios of the RNA copy numbers obtained from the two PCR analyses were calculated for each dilution , and their average value of 0 . 459 was set as a reference ratio ( S1D Fig ) . In the subsequent assays , the “3’:5’-end ratio” was determined by normalizing the ratio of the HCV RNA copy numbers obtained in the two PCR analyses , with the reference ratio . Total RNAs isolated from cells infected with HCVcc , JFH-1 or J6/JFH-1 ( whole cell ) as well as their culture supernatant ( whole sup ) , were quantified by using 5’- and 3’-end specific qRT-PCRs . In whole cell preparations , an excess amount of 5’-end HCV RNA was observed compared to 3’-end HCV RNA . The 3’:5’-end ratios in cells infected with JFH-1 and J6/JFH-1 ( whole cell ) were 0 . 158 and 0 . 076 , respectively , which were markedly lower than the ratios in the whole sup ( for both JFH-1 and J6/JFH-1 ) ( Fig 1A ) . The supernatants collected from both the HCVcc-infected cells as well as the cell extracts were subjected to density fractionation followed by quantification of the 5’- and 3’-ends of HCV RNA and determination of the infectivity of each fraction ( Fig 1B ) . The distribution of 3’- and 5’-end viral RNA in fractions of the culture supernatant showed a similar pattern , with the highest RNA levels observed in the fraction with the highest infectivity ( Fig 1B , upper ) . As for the cell-derived fractions , the distribution pattern of the 3’-end RNA differed considerably from that of the 5’-end RNA; while the former was mainly detected in infectious fractions , the latter was broadly distributed throughout the fractions tested ( Fig 1B , lower ) . The 3’:5’-end ratios in density fractions with the highest infectivity from the JFH-1 ( Fig 1B , upper ) and J6/JFH-1 ( S2A Fig ) cultures were significantly higher than those observed in the whole supernatant ( Fig 1A ) . In addition , 3’:5’-end ratios calculated from the data with normalization shown in Fig 1B correlated positively with infectivity ( supernatant; r = 0 . 78 , intracellular; r = 0 . 72 ) ( Fig 1C and S2B Fig ) . The fractions were categorized as high or low infectious groups based on the median value of the infectivity of the fractions . The mean value of 3’:5’-end ratios was significantly higher in high infectious fractions ( HI ) than in low infectious fractions ( LI ) ( Fig 1D ) . HCV RNA in infected cells was assessed by Northern blotting with a 5’ UTR anti-sense RNA probe . In addition to a major band of viral RNA of around the genome length , considerable signals at a size smaller than 0 . 5 kb were observed in cells infected with HCVcc ( S3 Fig ) . Combined with the data of the 3’:5’-end ratios in cells , these results suggested that the highly abundant 5’-end containing HCV subgenomes found in the virus-infected cells were negatively selected during the late steps of the viral life cycle . Based on the above findings that the 3’:5’-end ratios of HCV RNA were higher in the culture supernatant , particularly in highly infectious density fractions , relative to those in cells of HCVcc cultures , we deduced that a cis-acting sequence , possibly a region containing the 3’ side of the HCV genome , may be involved in encapsidation of the viral RNAs . Although coupling of genome replication and encapsidation has been widely reported among positive-strand RNA viruses , a replication-coupled packaging mechanism for HCV has not been identified to date . Previous studies suggested that HCV potentially employs a virion assembly process which is independent of RNA replication [11] , [12] . We thus assessed how HCV RNA replication influences the efficiency of virus assembly by using a trans-packaging system , which produces trans-complemented HCV particles ( HCVtcp ) , as previously described [13] . To produce HCVtcp , Huh7 . 5 . 1 cells were co-transfected with pHH-based plasmids expressing a replicative ( pHHSGR-JFH1/Gluc; WT ) or non-replicative ( pHHSGR-JFH1/Gluc/GND; GND ) JFH-1 subgenome and a plasmid pCAG/C-NS2 encoding Core , E1 , E2 , p7 and NS2 proteins . HCVtcp was inoculated into naïve Huh7 . 5 . 1 cells and transduced HCV RNA was determined ( Fig 2A ) . Although transduced HCV RNA in the cells inoculated with culture supernatants from GND- and Core-NS2-expressing cells was about 10-fold lower than that from WT-expressing cells , the transduced RNA levels were sufficient for detection ( Fig 2B ) . Transduced RNAs were detected with pre-treatment of the inoculum with benzonase , which proved that the transduced RNAs were introduced by nuclease-resistance structures ( S4 Fig ) . Anti-CD81 antibody clearly blocked RNA transduction via inoculation of culture supernatants from GND- and Core-NS2-expressing cells . In addition , Huh7-25 cells that are not susceptible to HCV entry due to a lack of CD81 expression [14] were not transduced , confirming the production of HCVtcp with a replication-defective JFH-1 genome ( Fig 2C ) . To test for pHH-based constructs for which it might be feasible to later analyze by mutational scanning to identify cis-elements of the HCV genome packaging , we removed the 5’- or the 3’ UTR of the GND subgenome . It was noted that production of HCVtcp dramatically decreased , close to the background level , when the 3’ UTR of pHHSGR-JFH1/Gluc/GND was deleted . Only a limited influence on HCVtcp production was observed by deletion of the 5’ UTR ( Fig 2D upper ) . Deletion of either UTR did not impair RNA stability ( Fig 2D , Northern blotting ) . Comparable expression of the mutant subgenomes and HCV structural protein was detected in the producer cells ( Fig 2D , middle and lower graphs ) . The combined results strongly suggested that replication of HCV RNA is important for efficient virus production but is not indispensable for viral assembly . Deletion of the UTR regions in the pHH-based construct did not essentially affect RNA stability . Thus , this trans-packaging system based on a replication-defective subgenome , which is sensitive enough to detect relatively low levels of HCVtcp , can be applied to investigate cis-acting signals for encapsidation of HCV RNA , independent of RNA replication . To investigate molecular mechanism ( s ) of encapsidation of HCV RNA , we determined the interactions of Core with a variety of HCV RNA fragments that have conserved sequences among HCV isolates and that can potentially fold into highly ordered stem-loop structures ( Fig 3A and S5 Fig ) . The interaction strength between in vitro synthesized Core tagged with an N-terminal FLAG and a series of biotinylated HCV RNAs was assessed by using AlphaScreen [15] . Core interacted with all of the RNA fragments tested . Notably , the highest assay signal was observed with the entire 3’ UTR ( 3’ UTR ) , compared to the signals obtained with the 5’ UTR , nt 9038–9257 ( SL9038-SL9198 ) , the 3’-163 nt of NS5B ( CRE ) , CRE plus the 3’ UTR ( CRE3’UTR ) , the RNA deleting 3’ X tail of CRE3’UTR ( CREVSL ) or the 268 nt-RNA fragment derived from the cellular hnRNPU gene ( Fig 3A and 3B , left ) . CRE is known to form a long distance kissing-loop structure with SLII of the 3’ UTR , which is crucial for viral RNA replication [8] . Addition of CRE to the 3’ UTR ( CRE3’UTR ) resulted in reduction of the Core-3’ UTR interaction . Core interaction with RNA elements within the 3’ UTR was then further assessed ( Fig 3A and 3B , right ) . Neither single- ( SLI , SLII , and SLIII ) nor double- ( SLI&II , SLII&III ) stem-loop structures , nor the 3’X region of the 3’ UTR exhibited efficient interactions with Core , compared to interaction of the entire 3’ UTR . Thus , it is likely that Core preferably binds to a pocket or to surface RNA structures composed of virtually the entire 3’ UTR . Indeed , deletion of the entire 3’ UTR from pHHSGR-JFH1/Gluc/GND caused a dramatic decrease in HCVtcp production ( Fig 2D ) . These findings encouraged us to test whether the 3’ UTR of HCV is sufficient to allow packaging of a foreign RNA sequence into HCVtcp . The HCV 3’ UTR sequence derived from JFH-1 ( genotype 2a ) or H77c ( genotype 1a ) isolates was inserted into a reporter plasmid , p/EmGFP , downstream of the coding region of Emerald Green Fluorescent Protein ( EmGFP ) , yielding p/EmGFP-3’UTR ( S6A Fig ) . Huh7 . 5 . 1 cells were transfected with p/EmGFP or p/EmGFP-3’UTR together with pCAG/C-NS2 and a plasmid encoding the NS3-5B polyprotein , pCAG/NS3-5B ( S6A Fig ) . We found that the EmGFP-3’ UTR but not the EmGFP sequence was packaged when C-NS2 and NS3-NS5B polyproteins were simultaneously supplied ( Fig 3C , upper ) . HCVtcp packaged with the EmGFP-3’ UTR was not produced in the absence of NS3-NS5B expression , demonstrating the involvement of NS proteins in HCV assembly . The 3’ UTR sequences derived from JFH-1 and H77c supported a comparative level of encapsidation of EmGFP RNA into HCVtcp ( Fig 3C , N/G3J VS N/G3H ) . Cell entry of the EmGFP-3’ UTR-packaged HCVtcp was blocked by anti-CD81 antibody ( Fig 3D ) . EmGFP , Core and NS5A expression , as well as EmGFP RNA levels in producer cells , were confirmed ( S6B and S6C Fig and Fig 3C , lower ) . The transduced RNA levels were sufficient for detection with pre-treatment of the inoculum with benzonase ( S6D Fig ) . Using this strategy , we tested whether the 5’ UTR of HCV supported encapsidation of EmGFP RNA . Although the 5’ UTR sequence appeared to support RNA encapsidation to some extent , the efficiency was significantly lower than that of the 3’ UTR ( Fig 3E ) . A moderate increase in transduced EmGFP RNA was detected when it was flanked with both the 5’ UTR and the 3’ UTR , compared to the version followed with the 3’ UTR alone ( Fig 3E ) . Competitive binding assays demonstrated that the 5’ UTR fragment did not compete with the 3’ UTR for Core-binding ( S6E Fig ) . When considered together with the HCVtcp production shown in Fig 2D and the in vitro binding data shown in Fig 3B , these findings suggested that , although both UTRs may be involved in HCV genome packaging , the packaging function of the 3’ UTR is higher compared to that of the 5’ UTR . To further investigate the microenvironment of foreign RNA ( Renilla luciferase ( Rluc ) RNA in this setting ) packaging into HCVtcp , we assessed the localization of HCV proteins in producer cells using confocal laser scanning microscopy ( Fig 4 ) . Huh7 . 5 . 1 cells were co-tranfected with pCAG/NS3-5B , pRluc ( R ) or pRluc-3’UTR ( R3 ) , together with pCAG/C-NS2 ( C-NS2 ) or an empty vector , and immunostained for NS5A and lipid droplets ( LD ) or NS5A and Core after 48 hr . The co-localization of NS5A with LD or Core was further analyzed by quantifying the Pearson’s correlation coefficient ( PCC ) [16] and intensity correlation quotient ( ICQ ) [17] . The degree of co-localization of NS5A and LD was significantly increased when Rluc-3’ UTR and Core-NS2 were co-expressed ( R3 , C-NS2 ) , compared to expression of Rluc ( R ) alone , Rluc plus Core-NS2 ( R , C-NS2 ) , or Rluc-3’ UTR ( R3 ) ( Fig 4A , left and middle , and Fig 4B ) . NS5A-Core co-localization was also significantly increased in cells expressing Rluc-3’ UTR with Core-NS2 , compared to cells expressing Rluc with Core-NS2 ( Fig 4A , right and Fig 4C ) . These results suggested that the 3’ UTR sequence facilitates the interaction between Core and NS5A at or around LD in HCVtcp-producing cells . These combined findings demonstrated that the 3’ UTR was more capable of interacting with Core and directing packaging of foreign RNA into HCV particles than the 5’ UTR . The 3’ UTR functioned as a cis-acting signal for efficient encapsidation and , not only Core-NS2 , but also NS3-NS5B polyproteins were involved in trans-complementing viral assembly as a means to encapsidate foreign RNA in combination with the HCV 3’ UTR . To further address the role of the 3’ UTR in packaging of HCV RNA , we introduced a series of deletion mutations into the 3’ UTR of the non-replicative subgenome pHHSGR-JFH1/Gluc/GND ( GND ) , and examined the effects of these mutations on the production of HCVtcp . The resultant mutants were ΔSLI , ΔSLII and ΔSLIII , in each of which one stem-loop of the 3’ X region was deleted; ΔVSL , in which the variable region and the poly ( U/UC ) tract were deleted; and ΔSLII&SLIII , ΔSLI&SLII , and Δ3’X , in each of which two or three stem-loops of the 3’ X region were deleted ( Fig 5A ) . Comparable levels of mutant and GND RNAs in the producer cells demonstrated comparable stability of the RNAs expressed ( S7 Fig ) . All of the mutants showed equivalent expression of Gluc to GND ( Fig 5B , middle ) , which also indicated that a similar level of RNA was expressed under the pol I promoter . The expression level of Core in the producer cells was also comparable ( Fig 5B , lower ) . As described above , deletion of the whole 3’ UTR ( Δ3’UTR ) markedly impaired HCVtcp production by 85% compared to that of GND . The mutants Δ3’X , ΔSLI&II , ΔSLII&III and ΔVSL reduced HCVtcp production by 68% , 55% , 33% and 27% , respectively . Among mutants with deletions of single stem-loop structures , deletion of either SLI or SLII , but not of SLIII , reduced HCVtcp production to some extent ( Fig 5B , upper ) . These results , together with the data of RNA-Core binding ( Fig 4B ) , suggested that the entire 3’ UTR of the HCV genome is important for efficient encapsidation . The 3’ X region , particularly SLI and SLII in this region , was indispensable for efficient encapsidation , while the variable region and poly ( U/UC ) stretch were also possibly involved in packaging . To verify the importance of SLI and SLII in these processed , effects of mutations in these regions ( Fig 5C ) on Core binding and HCVtcp production were determined . The interaction of the 3’ UTR with Core in vitro was significantly reduced when nucleotides in the loop regions of SLI and SLII were replaced with their paired counterparts ( LIM , LIIM , LI&IIM ) ( Fig 5D , upper ) . Conversely , the interactions of the 3’ UTR with Core were maintained when substitution mutations were introduced into the stem regions of SLI and SLII ( STIM and STIIM ) ( Fig 5D , lower ) . Accordingly , mutations in the loops but not in the stems of SLI and SLII led to a marked decrease in HCVtcp production ( Fig 5E ) . When LI&IIM mutations were engineered into the full-length JFH-1 genome , both genome replication and infectious virus production were markedly impaired as expected . Notably , a greater level of reduction of virus production than of genome replication was detected ( S8 Fig ) . On day 3 post-transfection with pHH-JFH1 ( WT ) and pHH-JFH1-X-LM ( X-LM; JFH-1 with LI&IIM mutations ) , the viral RNA level was around 5-fold lower in the cells replicating the mutant JFH-1 than in the cells with wild-type . In contrast , the mutant JFH-1 produced a 25-fold lower level of infectious particles than the wild type . This finding demonstrated that the apical sequences in stem-loops I and II are not only involved in HCV genome replication but also affect step ( s ) in virion assembly , and this result is consistent with the findings regarding Core-interaction and HCVtcp production . Based on the collected data , we propose a model whereby encapsidation of the HCV genome is potentially triggered by direct contact of Core with loop regions in the 3’ X region of the 3’ UTR , while other RNA structures of the 3’ terminus act as brace backbones to support this interaction and/or nucleocapsid formation .
Evidence regarding the virion assembly of HCV has accumulated over the past years; however , the detailed mechanisms responsible for incorporation of viral genomes into progeny virions remain largely unclear . A large proportion of 5’ end subgenomic HCV RNA with heterogeneous termination sites has been reported in the livers of hepatitis C patients [9 , 10] , consistent with a phenomenon found in viral-replicating cells . The progressive increase in 3’:5’-end ratios of HCV RNA in highly infectious density fractions and in whole-culture supernatant relative to viral replicating cells indicated selective packaging of integrated genomes into infectious particles . In the light of these results , we speculate that HCV may use a packaging-signal-dependent encapsidation pathway . Known packaging signals for RNA viruses are usually located in highly structured and conserved regions . The most structured and conserved sequences of HCV are the untranslated regions at either end . The findings that HCV RNAs containing a 3’-end side were enriched through particle assembly encouraged us to determine whether the 3’-end region of the HCV genome contains cis-packaging elements . Formation of a pseudoknot structure through a kissing interaction between the 3’ X region of the 3’ UTR and CRE within the NS5B region is essential for viral RNA replication [18] . Taking into account that the signals required for encapsidation might directly overlap with those required for replication , adoption of an approach by which the packaging process can be uncoupled from replication/translation is required for investigation of the process of encapsidation . Here we demonstrated that production of virions with both replicative and non-replicative subgenomes was possible in Huh7 . 5 . 1 cells . Notably , either 3’- or 5’-UTR deletion impaired HCVtcp production , but neither of these deletions perturbed RNA stability . Deletion of the 3’ UTR resulted in a decrease in transduced HCV RNA to the background level , while deletion of the 5’ UTR caused a minor reduction . It may be possible that the 5’ UTR is also somehow involved in efficient packaging of the HCV genome . In general , initiation of selective encapsidation of a viral genome requires recognition of a packaging signal by the nucleocapsid protein . Several conserved RNA structures across the HCV genome ( the 5’ UTR , SL9038-SL9198 , CRE in the NS5B-coding sequence , and the 3’ UTR ) were tested for in vitro interaction with Core . Interestingly , we found that the interaction of CRE in the NS5B-coding sequence with Core turned out not to be as strong as the interaction of the 3’ UTR with Core . When the 3’ UTR was linked to CRE ( CRE3’UTR ) , the strength of its interaction with Core was lower than that of the 3’ UTR alone . It was likely that access of Core to the 3’ UTR was implicated in the conformational switch of RNA-RNA interactions such as the kissing-loop structure formed via the interaction between CRE and the 3’ X region . Previous observations suggested that structural rearrangements within the 3’ end region of the HCV genome are important for the regulation of switching between different steps of the HCV lifecycle [19–21] . Furthermore , the HCVtcp system was modified by introducing a reporter gene cassette ( EmGFP-3’ UTR ) for encapsidation . Based on the results of the two HCVtcp systems and the RNA-Core interaction assay , we concluded that the 3’ UTR of the HCV genome functions as a cis-acting element for RNA packaging . The available evidence suggests that the packaging signal for the HCV genome is not contained within the 5’ UTR [22] . Our experimental approaches indicated that the 5’ UTR potentially supports packaging of foreign RNA . However , it does so significantly less efficiently than the 3’ UTR , which was consistent with our observation of impairment of HCVtcp production by deletion of the 3’- or the 5’ UTR . The packaging efficiency of the 3’ UTR was not compromised by the 5’ UTR when these two RNA structures were simultaneously supplied in separate cassettes . It appears that the SLI and SLII regions of the 3’ UTR play a key role in HCV encapsidation . Furthermore , mutations in the loops of SLI or SLII , which were predicted not to affect RNA secondary structures , led to strong negative effects on both Core-binding and trans-packaging efficiency . Nevertheless , mutations in the stem regions did not influence Core-binding or HCVtcp production . The predicted structures ( by RNAstructure software ) of STIM and STIIM showed that , even when the substitutions in the stems dramatically changed the RNA structures , there were no major changes in the loop area . Thus , the stem mutants still contained the loop motifs of the wild-type RNA ( S9 Fig ) , which possibly explained the unchanged interaction with Core . Based on the combined data , we deduced that the entire 3’ UTR is needed for efficient packaging , while the apical sequences of SLI and SLII are crucial for this process . When the same mutations of apical sequences were engineered into full-length JFH-1 , greater reduction of virus production than that of genome replication was detected , which supported the findings that these sequences are also involved in virion formation independent of replication . Therefore , in HCV , sequence motifs located in the loops of multiple stem-loop RNA structures were crucial for encapsidation , similar to packaging signals of hepatitis B virus and alphaviruses [23 , 24] . In the setting with EmGFP-3’UTR as a gene cassette for HCVtcp production , it is of interest that co-expression of NS3-NS5B together with Core-NS2 was a prerequisite for packaging of the reporter gene . Presumably , in the HCVtcp system used , the nonstructural proteins did not function as replication machinery but rather contributed to creating the subcellular environment required for virion assembly . In addition to viral structural proteins and p7 , all NS proteins contribute to the production of infectious HCV particles [25] . NS2 and p7 are crucial for virus assembly and release [26–31] and can be trans-complemented[32] . NS3 , NS4A , NS4B , NS5A , and NS5B are indispensable for HCV genome replication and virus assembly [33–36] . Although NS4B and NS5A can be trans-complemented for replication [31 , 37] , it has not been determined whether NS3-NS5B can be trans-complemented for virion production . Here , for the first time , we demonstrated that all HCV proteins could be supplied in trans for virion packaging in a replication-defective HCVtcp system . Impaired trans-packing of virions with mutations in domain III of NS5A indicated direct involvement of this domain in encapsidation . The 3’ UTR of the HCV genome has been found to activate an IKK-α-dependent pathway that induces lipogenic genes and enhances Core-associated LD formation to facilitate viral assembly [38] . We observed that association of NS5A with LD and Core was enhanced by co-expression of the 3’ UTR . It is likely that , in addition to its direct role as a packaging signal , the 3’ UTR induces the host cell environment required for facilitation of HCV morphogenesis . Functional coupling of RNA packaging and replication have been reported for several positive strand RNA viruses [1–3] . One may hypothesize that a replication-coupled packaging mechanism might permit efficient access of the nucleocapsid protein so that it can interact with progeny HCV RNA . However , this coupling mechanism may result in competition of Core and the replicase for RNA binding . Thus , temporal recruitment of Core to LDs avoids such competition [39] . It was proposed that the newly synthesized HCV genome RNA , coupled with NS5A , is released from the replication complex-containing membrane vesicles , and recruited to the surface of LDs or their associated membranes [33 , 35 , 40 , 41] where nucleocapsids are assembled . In this study , we acquired comprehensive knowledge regarding HCV RNA associated with infectious particles and identified the packaging signal of HCV virus encapsidation . The findings discussed here will help to decipher the complicated process of the early phases of virion assembly of HCV . Our work provides a rational framework for refining the molecular mechanisms regulating the HCV lifecycle and potentially enables determination of mechanisms that govern other related positive strand RNA viruses .
The in vitro synthesized viral genomic RNA was prepared with a MEGAscript T7 kit ( Life technologies ) , using a linearized HCV cDNA clone as the template . The genome-size RNA was purified from agarose gel after formaldehyde denaturing agarose gel electrophoresis of the synthesized RNA . Uniformity of the IVT-F RNA was assessed by Agilent 2100 Bioanlyzer in combination with the Agilent RNA 6000 Nano Kit . Quantification of HCV RNA was performed by qRT-PCRs using TaqMan EZ RT-RNA Core Reagents ( Applied Biosystems ) . HCV RNA quantifications were done by targeting either 5’ UTR or NS5B , in two ( q ) RT-PCR sets with respective RNA standards . The primers and probes used for quantifying the 5’ UTR of HCV have been described previously [42] . To set up the quantification of HCV RNA targeting the 3’ end of the genome , the 3’-end sequence ( nt 8001–9678 ) of JFH-1 isolate , including partial NS5B and 3’ UTR was subjected to screen the best primers and probes by IDT Scitools RealTime PCR ( Integrated DNA Techonologies ) . Initially , two probes in the NS5B region and one probe in the 3’ UTR region were estimated . The working efficiencies of these probes were compared and the one with best efficiency was picked for quantification of HCV RNA in name of 3’ end qRT-PCR . Thus , the primer/probe set located at NS5B ( nt 8057–8196 ) , a forward primer 5'-CAA ACA CCA ATT CCC ACA ACC -3' and reverse primer 5'-TCA TAG AGG GCC ATT TTC TCG -3’ , and the probe 5'-/56-FAM/AC CAG CTC G/ZEN/C CTC ATC GTT TAC C/3IABkFQ/-3' , were used . The assay for quantification of HCV RNA at 5’ UTR and NS5B was performed as follows: for each sample , RNA was mixed with 3 mM of MnOAc , 10 mM of each dNTP , 10 μM of forward and reverse primer , 10 μM of TaqMan probe , 2 . 5 U of rTh polymerase , 0 . 5 U of Amp Erase UNG , in the TaqMan EZ buffer scale to 15 μl with double distilled water . PCR run program: starting with reverse transcription at 50°C for 1 min , 60°C for 50 min and 95°C for 5 min , then followed by 50 cycles of target amplification , denaturation at 94°C for 15 s , annealing at 55°C for 10 s and extension at 72°C for 1 min . A primer/probe set for qPCR targeting EmGFP gene was selected from validated Assays-on-Demand products ( RIKAKEN ) . Total cellular RNA was extracted with ReliaPrep RNA purification kit ( Promega ) and treated with TURBO DNase ( Life technologies ) , followed by cleaning up with a Nucleospin RNA clean-up kit ( TaKaRa Bio ) . RNA from supernatant or density fractions was extracted with a SepaGene RNA extraction kit ( Edia Japan ) according to manufacturer’s instructions . HCVtcp was generated by cotransfection with pCAG/C-NS2 and pHHSGR-JFH1/Gluc essentially as previously described [13] . To produce HCVtcp carrying EmGFP RNA , cells were cotransfected with pCAG/C-NS2 , pCAG/NS3-NS5B and either p/EmGFP , p/EmGFP-J3UTR , p/EmGFP-H3UTR , p/5’UTR-EmGFP-J3’UTR or p/5’UTR-EmGFP using TransIT-LT1 ( Mirus Bio ) . At 72 hr post-transfection , cultured media and cells were harvested and assessed for infectivity by inoculating naïve Huh7 . 5 . 1 cells and expression of viral proteins by Western blotting , respectively ( see S1 Text for details ) . Culture medium containing HCVtcp was filtered through a 0 . 45 μm Sterivex filter unit ( Millipore ) and 20 μl of it was used to measure Gluc activity . The remaining filtered medium was used to inoculate Naïve Huh7 . 5 . 1 cells maintained at 37°C with 5% CO2 . After 3 hr inoculation , cells were washed with PBS and cultured 12 hr with fresh growth medium for RNA extraction . Total RNAs of transfected Huh7 . 5 . 1 cells were isolated with Tri-reagent ( Sigma-Aldrich ) , at 72 hr post-transfection . The isolated RNAs were treated with TURBO DNase ( Life technologies ) , followed by cleaning up with a Nucleospin RNA clean-up kit ( TaKaRa Bio ) . The resulted RNAs were then analyzed by electrophoresis in a 1 . 5% agarose-2 . 2 M formaldehyde gel , followed by Northern hybridization by using a DIG Northern starter kit ( Roche ) according to the manufacturer’s instructions . DIG-labeled anti-sense RNA probe complementary to the positive strand of NS5B ( nt 7951–8476 ) was used to detect HCV RNA . Biotinylated RNA fragments were synthesized using a MEGAscript T7 kit ( Life technologies ) or provided by Sigma Aldrich . The size and stability of RNA fragments synthesized with MEGAscript kit were checked by agarose gel electrophoresis . RNA fragments purchased from Sigma Aldrich were provided in 0 . 05 μmole scale , HPLC grade purity . FLAG-tagged HCV Core and its mutants were synthesized using a WEPRO in vitro translation kit ( CellFree Sciences ) using pEU/core/wt as a template . The RNA-Core interaction assays were carried out in a final reaction volume of 25 μl by adding each assay component to the following final concentrations: 0 . 32 nM , 1 . 6 nM , or 8 nM of biotinylated RNA , 20 nM of in vitro synthesized Core protein , 0 . 1% BSA ( w/v ) , 40 U RNase inhibitor , 20 μg/ml anti-FLAGm2 acceptor beads , 20 μg/ml streptavidin donor beads , 2 . 5 μl of 10x AlphaScreen assay buffer ( PerkinElmer ) , followed by 90 min of incubation at 25°C and then subjected for determination of interactions by AlphaScreen signals ( photon counts at 630 nm/s ) . AlphaScreen signals were detected on an EnSpire plate reader ( PerkinElmer ) . The culture supernatant of Huh7 . 5 . 1 cells infected with HCVcc was concentrated by ultrafiltration with a centrifugal filter device ( Amicon ) , and filtered with a 0 . 22 μm Sterivex filter unit ( Millipore ) . To release intracellular virus , cell pellets were re-suspended with Dulbecco’s modified Eagle’s medium ( DMEM ) containing 10% fetal bovine serum and subjected to four cycles of freezing and thawing , followed by centrifugation at 2 , 400 x g for 10 min to remove cell debris . Supernatant- or cell-derived samples were then layered on top of 30 ml of a 10% to 50% Opti-prep density gradient medium ( Sigma-Aldrich ) prepared in DMEM and centrifuged in a Beckman SW41Ti rotor ( Beckman ) at 25 , 000 rpm for 16 hr at 4°C . Fractions of 1 ml were collected from the bottom of each tube . Immunostaining of LD , NS5A and Core were performed as previously described [41] . Subcellular localization of HCV proteins was observed using an Olympus FLUOVIEW FV1000 confocal laser scanning microscope . Co-localization of NS5A with Core or LD was quantified via Pearson’s correlation coefficient ( PCC ) [16] and intensity correlation quotient ( ICQ ) [17] using ImageJ software ( http://imagej . nih . gov/ij/ ) . | Although cell culture systems provide a powerful tool for deciphering the life cycle of the hepatitis C virus ( HCV ) , the mechanisms of encapsidation of the viral genome into infectious particles remain to be uncovered . The HCV genome is a positive RNA with one single reading frame flanked by 5’- and 3’ untranslated regions ( UTRs ) . Thus far , there is no direct evidence that HCV employs a packaging-signal dependent- or replication-coupled mechanism of encapsidation of its genome . The possible overlap of RNA sequences that function in RNA replication with those that function in encapsidation may present an obstacle to investigation of the cis-elements for RNA packaging . In this study , we characterized the properties of HCV RNAs in a cell culture system by determining their integrity in virus-replicating cells and in culture supernatants , and we found that over-distributed 5’-subgenomes were negatively selected during virus assembly in the cells . Using trans-packaging systems with replication defective subgenomes , we identified the 3’UTR as a cis-acting element that was sufficient for packaging of not only a HCV subgenome but also a foreign gene into infectious particles . Mutagenesis analyses , together with an in vitro binding assay with Core demonstrated that , whereas the best encapsidation occurs with the entire 3’ UTR , the loop sequences of the 3’ X region appear to be essential for encapsidation . Our work opens new perspectives for understanding the molecular mechanisms that regulate the HCV life cycle and potentially paves a way to a new anti-viral therapy . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"and",
"health",
"sciences",
"nucleic",
"acid",
"synthesis",
"pathology",
"and",
"laboratory",
"medicine",
"3'",
"utr",
"hepacivirus",
"pathogens",
"messenger",
"rna",
"rna",
"extraction",
"microbiology",
"viruses",
"rna",
"viruses",
"untranslated",
"regions",
"rna",
"synthesis",
"extraction",
"techniques",
"chemical",
"synthesis",
"research",
"and",
"analysis",
"methods",
"rna",
"structure",
"medical",
"microbiology",
"microbial",
"pathogens",
"hepatitis",
"c",
"virus",
"hepatitis",
"viruses",
"viral",
"replication",
"molecular",
"biology",
"biosynthetic",
"techniques",
"biochemistry",
"rna",
"nucleic",
"acids",
"flaviviruses",
"encapsidation",
"virology",
"viral",
"pathogens",
"5'",
"utr",
"biology",
"and",
"life",
"sciences",
"organisms",
"macromolecular",
"structure",
"analysis"
] | 2016 | Involvement of the 3’ Untranslated Region in Encapsidation of the Hepatitis C Virus |
Drug discovery initiatives , aimed at Chagas treatment , have been hampered by the lack of standardized drug screening protocols and the absence of simple pre-clinical assays to evaluate treatment efficacy in animal models . In this study , we used a simple Enzyme Linked Aptamer ( ELA ) assay to detect T . cruzi biomarker in blood and validate murine drug discovery models of Chagas disease . In two mice models , Apt-29 ELA assay demonstrated that biomarker levels were significantly higher in the infected group compared to the control group , and upon Benznidazole treatment , their levels reduced . However , biomarker levels in the infected treated group did not reduce to those seen in the non-infected treated group , with 100% of the mice above the assay cutoff , suggesting that parasitemia was reduced but cure was not achieved . The ELA assay was capable of detecting circulating biomarkers in mice infected with various strains of T . cruzi parasites . Our results showed that the ELA assay could detect residual parasitemia in treated mice by providing an overall picture of the infection in the host . They suggest that the ELA assay can be used in drug discovery applications to assess treatment efficacy in-vivo .
Chagas disease ( CD ) is prevalent across various countries in Central and South America , with 28 million people at risk of getting infected [1] , [2] . The etiological agent of CD , Trypanosoma cruzi , has two life cycle stages in the infected host , an extracellular trypomastigote form that circulates in blood of infected individuals , and an intracellular amastigote form that is present in the infected tissues and organs , such as the heart [3] , [4] . Parasite infection and disease progression typically presents an acute phase characterized by blood parasitemia followed by a lifelong chronic phase , when circulating parasites can no longer be detected by direct blood examination [5] . Only two drugs , Benznidazole ( Bz ) and Nifurtimox have been the mainstay of treatment since the 1960–70s [6] , [7] . However , these drugs produce multiple side effects leading frequently to early termination of treatment [8] . Additionally , the presence of intracellular amastigotes in the tissues and organs , where drug levels may not reach therapeutic concentrations , act as reservoirs of infection in the host [9] , [10] . It has been estimated that for chronic individuals>80% of treated patients do not demonstrate parasitological cure and in the absence of drug therapy , or in immuno-compromised patients , recurrence of parasitemia is common [11] , [12] . In light of these issues associated with CD therapy , new and better drugs need to be developed . Chagas drug discovery initiatives have been hampered by the lack of standardized drug screening protocols , for example , pre-clinical assays that can be used to evaluate cure in animal models [13] . As trypomastigote levels can fluctuate in blood , detection of T . cruzi amastigote DNA by PCR in organs of infected drug-treated mice is a definitive end-point assay for determining cure [14] , [15] . Immunosuppressing drug treated animals and allowing sub-patent infection to present itself as blood parasitemia , detectable using microscopy or PCR is another end-point assay indicating drug treatment failure [16] . Recently , several drugs , such as Posaconazole , VNI and Fexinidazole have been shown to cure mice using these methods [14] , [15] , [17] . However , for successful drug discovery , the ability to rapidly screen several drugs or their derivatives in-vivo for structure activity relationship studies is critical . Thus , due to the technical complexities associated with tissue PCR and immunosuppression protocols that require euthanization , these assays are far from ideal for pre-clinical drug discovery efforts [13] . These technical and other economic considerations have resulted in a significant lack of investment by the pharmaceutical industry in developing new drugs for neglected tropical disease such as Chagas disease [18] . We have recently reported an aptamer-based , non-serological , non-PCR assay , to detect T . cruzi biomarkers circulating in the blood of infected mice [19] . Aptamers are short nucleic acids , such as DNA or RNA , that are selected from a random library to bind specific targets , using a method called Systematic Evolution of Ligands by Exponential Enrichment ( SELEx ) [20] . We utilized this method to develop aptamers against a complex target: the T . cruzi excreted secreted antigens ( TESA ) [19] . In the current report , we performed SELEx to isolate additional aptamers that could detect T . cruzi biomarkers in murine drug discovery models of CD .
Animal care protocol was approved by the Center for Biologics Evaluation and Research Animal Care and Use Committee ( Protocol ASP #2010-03 ) and experiments performed following NIH Animal Care and Use Committee guidelines . Five to seven week old female , Swiss and C57BL/6 , mice were infected with 5000 and 1000 blood stage T . cruzi trypomastigotes , respectively [13] , [19] , [21] . Swiss mice infected with the T . cruzi Colombiana strain were representative of the acute phase model of CD [13] . C57BL/6 mice infected with the T . cruzi Colombiana strain were representative of the chronic phase model of CD [21] . Parasite levels were determined using light microscopy obtained via tail vein bleed [19] , [22] . The myotropic Colombiana strain and the Y strain of T . cruzi were maintained by serial passage every 21 and 10 days , in Swiss mice , respectively . The Y , Colombiana and 0704 strains of T . cruzi were cultured in-vitro using 3T3 mouse fibroblast cells [19] , [23] . Infected 3T3 cell culture supernatant was collected and concentrated to obtain the TESA fraction [19] . Protein extracts from parasites unrelated to T . cruzi , such as Plasmodium falciparum infected RBCs , promastigotes of Leishmania donovani , Leishmania major and Leishmania infantum were obtained as described earlier [19] . Both infected and non-infected mice were injected intraperitoneally ( IP ) with 200 µl Benznidazole ( Bz , Sigma ) ( 100 mg per Kg body weight ) diluted in 50% DMSO in PBS , for 20 consecutive days [13] . After the 20 day treatment , mice were allowed to recover for 20 days and then bled and euthanized for collecting heart and skeletal muscle samples . Data presented are from one experiment , representative of three independent experiments , performed for each of the two drug treatment models . Aptamers were selected using an oligonucleotide pool Round 10 ( R10 ) described previously [19] . Conditions used for each additional round of SELEx , from round 11 ( R11 ) to round 21 ( R21 ) have been described in S1 Table . Briefly , 100 µM aptamer pool was incubated at 65°C for 10 minutes and allowed to refold at room temperature for 1 hour . The refolded aptamer pool was then diluted to 1 ml in PBS , filtered through a 0 . 22 µm nitrocellulose ( NC ) membrane . The filtrate was then used for SELEx as described in S1 Table . Aptamer pool obtained at the last round of SELEx was cloned and biotinylated monoclonal aptamers and pools produced as described earlier [19] , [23] . ELISA plates were coated with 50 µL of 2 . 5 µg protein/well of TESA , trypomastigote or epimastigote extracts or with mice plasma samples diluted 1∶200 in PBS [19] . Coated plates were blocked with 1% bovine serum albumin ( BSA ) in PBS . After discarding the blocking buffer , biotinylated RNA aptamers were added to each well . After 1 hour incubation the plate was washed thrice with PBS to remove unbound aptamers . Streptavidin-alkaline phosphatase was added to the wells and bound conjugate detected using 4-Methyllumbelliferyl Phosphate ( 4-MUP ) ( Liquid Substrate System , Sigma ) [19] . Fluorescence was measured at an excitation wavelength of 360 nm and emission wavelength of 440 nm , with a cutoff filter of 435 nm , using a Spectra Max , M5 , ( Molecular Devices ) [19] . The extraction of total genomic DNA from blood and tissue of animals infected with T . cruzi was performed using commercial kits , QIAamp DNA blood Mini kit , Qiagen and DNeasy Blood & Tissue Kit , respectively . Whole blood ( 50 µl ) was first lysed with 1400 µl of 5% Saponin ( Sigma ) prepared in PBS . The lysate was centrifuged at 5000×g and the DNA isolated from the pellet containing the unlyzed cells , including parasites . The PCR reactions were performed using 50 ng of DNA obtained from genomic tissue and 1 µl for DNA solution obtained from blood . The reaction mixture contained 0 . 4 µM of forward and reverse primers ( F: 5′-AGTCGGCTGATCGTTTTCGA-3′; R: 5′-AATTCCTCCAAGCAGCGGATA-3′ ) , 1× Premix Ex Taq ( 2× premix -Takara System ) and 0 . 3 µl of SYBR Green ( from a 100× stock , Invitrogen ) . For tissue PCR the genomic region encoding the IL12 p40 was utilized as the control to demonstrate equal amounts of purified DNA were used for all PCR amplifications . The genomic IL-12 p40-specific primers utilized were 5′-GTAGAGGTGGACTGGACTCC-3′ and 5′-CAGATGTGAGTGGCTCAGAG-3′ [21] . PCR was performed using the Bio-Rad CFX real time PCR machine with cycling conditions as follows: 95°C for 30 seconds; 95°C for 5 seconds and 64 . 2°C for 30 seconds , for 45 cycles . Amplification was immediately followed by a melt analysis to confirm specific amplification of parasite DNA [19] , [23] , [24] . Data was plotted and statistical analysis carried out using GraphPad PRISM . Unpaired t-test with Welch correction was performed with a 95% confidence interval and p-values calculated . The ELA assay cutoff was calculated from the formula: mean RFU of control group ( i . e . , the non-infected drug treated group ) + ( 2× standard deviation ) .
We have recently reported the development of an Enzyme Linked Aptamer ( ELA ) assay to detect circulating T . cruzi Excreted Secreted Antigens ( TESA ) in the blood of the infected mice [19] . Eleven additional rounds of SELEx were performed under more stringent conditions with the starting library of round 10 ( R10 ) , obtained from our previous report , to increase the signal to noise ratio of the ELA assay ( S1 Table ) [19] . During each subsequent round of SELEx , the RNA aptamer pool was first depleted of non-specific aptamers , termed negative SELEx , by incubating the library with non-infected mouse plasma , and protein extracts from other parasites such as promastigotes of L . donovani , L . major , L . infantum and P . falciparum infected red blood cells ( iRBC ) ( S1 Table ) . ELA assay data with 5′-biotinylated aptamer pools from rounds 1 , 5 , 10 , 14 , 16 , 18 and 21 indicated that , as the SELEx progressed , the proportion of TESA binding aptamers that constitute the pools from rounds R10 to R21 increased significantly ( Fig . 1A ) . Compared to the R10 pool , R21 pool showed almost 6 fold increase in TESA binding signal ( Fig . 1A ) . Aptamer pools R16 , 18 and 21 showed binding at concentration as low as 32 . 25 nM , suggesting that they contained aptamers with higher binding affinities compared to aptamers obtained from R10 . Dose response curve for the R21 pool showed saturable binding to TESA at concentrations ranging between 25 to 50 nM , thus indicating that the individual aptamers in this pool were high-affinity binders with minimal binding to the BSA control ( Fig . 1B ) . The R21 aptamer pool also showed saturable binding to trypomastigote extract and at lower levels to T . cruzi epimastigote extracts ( Fig . 1B ) . Comparison of ELA binding data indicated that aptamers at R21 gave a significantly higher signal with the infected mice plasma compared to those at R10 ( Fig . 1C ) . The signal to noise ratio ( RFU Infected/RFU Non-Infected ) improved from 1 . 3 for the R10 pool to 3 . 2 for the R21 pool . The higher stringency of SELEx conditions used from R10 to R21 improved the signal to noise ratio of the ELA assay , primarily due to enhanced binding characteristics of the R21 aptamer pool to plasma from T . cruzi infected mice ( Fig . 1C ) . The R21 aptamer pool was cloned and sequenced to identify individual aptamer sequences . Phylogenetic analysis indicated that the sequences had converged to 7 major families , with a majority of the clones ( >70% ) belonging to Family 3 ( S1 Fig . ) . The monoclonal aptamer sequences ranged from 72 to 145 base-pairs in length ( Fig . 2A ) . A representative clone from each family was PCR amplified and used to generate 5′-biotinylated aptamers [19] , [23] . All the selected biotinylated aptamers , Apt-1 , 29 , 71 , 74 , 75 , 77 and 79 , showed saturable binding to TESA but not to BSA which was used as a negative control ( Fig . 2B ) . These aptamers also showed saturable binding to T . cruzi trypomastigote extracts ( Fig . 3 ) . Individual aptamers demonstrated higher signals with TESA than with equivalent amounts of trypomastigote protein extract ( Fig . 3 ) . This may be due to the higher relative abundance of their targets in the TESA fraction compared to the trypomastigote lysate . RNA Fold software , indicated that the 7 aptamers selected from R21 folded in highly stable three dimensional structures , formed by the various stem-loop regions of the sequences , with negative ΔG values ranging from −17 . 2 kcal/mol to −32 . 2 kcal/mol ( S2 Fig . ) . A group of Swiss mice were infected with the Colombiana strain of the T . cruzi parasite and then treated with Benznidazole ( Bz ) . Parasitemia was followed by microscopy and drug treatment was started at 15 days post infection ( dpi ) with 100 mg/kg/day Bz for a total of 20 days ( Fig . 4A ) . Microscopy showed T . cruzi trypomastigotes in the blood of all the infected animals at 15 dpi . At 55 dpi the mice were sacrificed and blood , heart and skeletal tissues were collected for PCR and ELA assays to detect biomarkers . All seven aptamers , Apt-1 , 29 , 71 , 74 , 75 , 77 , and 79 , were able to detect significantly higher levels of parasite biomarkers in plasma of infected mice compared to the age matched non-infected control mice at 55 dpi using the ELA assay ( Fig . 4B–C and S3 Fig . ) . Further , all the aptamers , except Apt-1 , showed significantly higher levels of biomarkers in the infected drug treated group compared to the non-infected drug treated controls ( Fig . 4B–C and S3 Fig . ) . The non-infected drug treated group was utilized to detect any interference in the ELA assay by the drug alone and to establish the signal cutoff for the infected drug treated group ( mean +2×s . d . of the non-infected treated group ) . At 55 dpi , ELA assays with Apt-29 showed that the biomarker levels in 100% ( 9/9 ) of the infected drug treated mice were above the cutoff of the assay , suggesting that these animals could still be infected ( Fig . 4D ) . Apt-29 ELA assay was as good as blood PCR ( 100% positive ) at detecting mice that were still infected post drug treatment , but was significantly better than tissue PCR , as 11% and 89% of the infected treated group had detectable parasite DNA in the heart tissue and skeletal muscles , respectively ( Fig . 4D ) . Similar results were observed with Apt-77 ELA , with 89% ( 8/9 ) of the mice showing biomarker levels greater than the cutoff ( Figure 4C–D ) . To demonstrate that aptamers could also predict the outcome of drug treatment in a chronic phase model of CD , a group of C57BL/6 mice were infected with the Colombiana strain of T . cruzi . Drug treatment was started at 130 dpi with 100 mg/kg/day Bz for a total of 20 days until 150 dpi ( Fig . 5A ) . At 170 dpi , all mice were sacrificed; blood , heart and skeletal tissues were recovered and analyzed by ELA assays and PCR . All seven aptamers were able to detect significantly higher levels of parasite biomarkers in plasma of infected mice compared to the age matched non-infected control mice at 170 dpi using the ELA assay ( Figures 5B–C and S4 Fig . ) . Further , all the aptamers except , Apt-79 and 74 , also showed significant difference between the level of biomarkers in infected drug treated group compared to the non-infected drug treated controls at 170 dpi ( Figures 5B–C and S4 Fig . ) . At 170 dpi , 100% ( 13/13 ) of the benznidazole treated mice were ELA positive using Apt-29 and 71 ( Fig . 5D ) . In contrast , 46% ( 6/13 ) of the drug treated animals remained positive by blood PCR , 38% ( 5/13 ) by heart muscle PCR and 92% ( 12/13 ) by skeletal muscle PCR ( Fig . 5D ) . Although one drug treated mouse was negative by blood and tissue PCRs , the internal control IL12 p40 was amplified from all these specimens and therefore the absence of T . cruzi DNA amplification was not due experimental errors . In this animal Apt-29 biomarker level was significantly higher ( 8047 . 61 RFU ) compared to the assay cut-off ( 5509 RFU ) and thus it was interpreted as infected ( Fig . 5D ) . These results showed that Apt-29 and Apt-71 ELA assays were able to predict the failure of Benznidazole treatment accurately in 100% of the mice in this chronic mouse model . To determine whether the aptamer-based biomarker detection assay could be used as a universal tool for anti-T . cruzi drug screening in murine models , we tested plasma obtained from mice in the chronic phase of infection with the Y and 0704 strains of the parasite . ELA assays performed with Apt-29 , 71 and 77 , showed that biomarker levels in these infected mice were significantly higher than the non-infected controls ( Fig . 6 ) . For Apt-77 , although the biomarker levels in the Y strain infected mice were not statistically significant , a positive trend was observed ( Fig . 6C ) . This data suggests that , the aptamers bind conserved targets across various strains of T . cruzi , the Tulahuen , used for SELEx , the Colombiana used for the acute and chronic phase models , and the Y and 0704 strains shown here , and that ELA assays could be used as a testing tool for anti-T . cruzi drug screening development using various murine models .
One of the primary reasons for failure to develop new drugs for CD has been the lack of standardized end-point assays in animal models [13] . The lack of universally accepted reliable tests to assess parasite burden and clearance and the hindrance this places on the development and evaluation of potential new drugs , both in experimental animal models and humans is well documented [25] , [26] . In this study we describe a biomarker-based detection assay to assess treatment efficacy in murine drug discovery models for the screening of anti-T . cruzi drugs . The fact that aptamers bound to purified trypomastigote extract and that T . cruzi infected drug treated mice showed reduced levels of biomarkers , compared to the infected group alone , indicates that the targets of the aptamers were of parasite origin . Current methods to assess cure in mouse models of CD are based on detecting parasite DNA in tissues samples or to immunosuppress drug treated animals and allow sub-patent infection to manifest as blood parasitemia that is detectable by microscopy or PCR [14] , [15] . However , in the absence of immunosuppression , tissue PCR may give false negative results , for example , when sampling a part of infected tissue that may not have intracellular amastigotes . Additionally , various strains of parasites have varied tissue tropisms , particularly with different MHC restricted murine hosts [4] , [27] . To overcome some of these issues , luciferase expressing parasites have been developed as tools for drug discovery [28] , [29] . Recent data has demonstrated that T . cruzi parasites progressively migrate to , and infect different organs of the host , with the gastro-intestinal tract being the major reservoir of infection in the chronic phase [29] . Thus to demonstrate cure in these situations , all the tissues may have to be sampled by PCR , making such models laborious and impractical for drug screening purposes . Further , this limits drug screening to a few parasite strains that may be amenable to genetic manipulation . A biomarker-based assay , on the other hand , such as the one presented here , is based on the detection of parasite excreted secreted proteins in biological samples including serum or plasma . As parasite antigens secreted in different tissues , and transported via the interstitial fluids , will eventually be present in blood , the ELA assay provides a global picture of parasitemia in the host [30]–[33] . A positive ELA assay result would suggest that the treated animals continue to harbor T . cruzi parasites somewhere in the body thus indicating drug treatment failure . Drug discovery models in the acute phase have been focused primarily on demonstrating the trypanocidal effect of the drug measured by the reduction or absence of parasites in blood , also termed parasitological cure , or massive reduction in parasite loads that could prevent further pathological damage [14] . However , most Chagasic patients are already in the chronic phase before their disease status is established and treatment can be initiated . In this regard , the BENEFIT trial is being conducted to determine the efficacy of treatment in the chronic phase [34] . The aptamers described in this report are able to detect parasite biomarkers in both the acute as well as the chronic phase of the disease . As the animals do not need to be euthanized , a biomarker-based approach could yield continuous longitudinal data on the therapeutic effect of the drug . Moreover , ELA assays can be performed using standard ELISA equipment and do not require a highly controlled environment free of potential contaminating DNA , expensive reagents and complex instrumentation as PCR or flow cytometry . Another advantage of the aptamer-based assay is that these aptamers also detect targets in various strains of parasites , suggesting , that drug sensitive as well as drug resistant parasite strains , can be used in a biomarker-based drug screening protocol . It is worth noting here that the T . cruzi Colombiana strain is significantly more virulent and shows higher parasitemia in the mice strains used in this report compared to the Tulahuen or CL strain of the parasite , making it more difficult to cure [15] , [35] . Another application where the ELA assay to detect biomarkers could be of use is the development of vaccines for CD [36]–[38] . Biomarker presence , detected by ELA assays , in parasite challenged vaccinated animals could indicate that the immune response was not sufficient to control the infection . The need for a reliable method to evaluate the cure of Chagas is a major concern , particularly when a clinical trial is designed . Furthermore , due to ethical reasons , commonly used protocols , such as tissue PCR and immunosuppression , cannot be used as endpoint assays for human clinical trials . The identification and validation of biomarkers for cure would be required before conducting human trials . Towards this eventual goal , further studies are being conducted to establish the utility of the reported aptamers for detecting biomarkers in samples from chagasic patients . In conclusion , we have successfully developed aptamers that demonstrate high binding specificity and sensitivity to TESA . Moreover , we demonstrated the power of ELA assay to detect T . cruzi biomarkers in the plasma of infected mice and to predict drug treatment failure in both the acute and chronic phase murine models of Chagas disease . ELA assay can be used in drug discovery applications to assess drug efficacy in-vivo . | A marked decrease in incidence of Chagas Disease has been observed in the last decade achieved by vector control strategies; however , there are still geographical areas where the disease reaches endemic proportions . Due to high morbidity and disease burden , other avenues of Chagas control , such as vaccines and therapeutic agents need to be employed for comprehensive disease control and mitigation . As there are no vaccines available currently , two drugs ( Benznidazole and Nifurtimox ) have been the mainstay of treatment . However , these drugs produce multiple side effects and frequently lead to early termination of the treatment . In this study , we have successfully developed a new method to evaluate the presence of Chagas biomarkers in the plasma of infected drug treated mice . Our study shows that high biomarker levels in T . cruzi infected mice , after drug treatment , can indicate treatment failure . Our assay provides a global picture of parasitemia in the host and a positive result would thus suggest that the treated animals continue to harbor T . cruzi parasites somewhere in the body . This study provides a new method to test for T . cruzi infection and for assessing the effectiveness of treatment . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"tropical",
"diseases",
"biomarkers",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"drug",
"screening",
"parasitemia",
"protozoans",
"pharmaceutics",
"neglected",
"tropical",
"diseases",
"pharmacology",
"quantitative",
"parasitology",
"veterinary",
"science",
"veterinary",
"parasitology",
"drug",
"discovery",
"biochemistry",
"trypanosoma",
"cruzi",
"trypanosoma",
"chagas",
"disease",
"drug",
"research",
"and",
"development",
"biology",
"and",
"life",
"sciences",
"drug",
"therapy",
"organisms"
] | 2015 | Aptamer-Based Detection of Disease Biomarkers in Mouse Models for Chagas Drug Discovery |
Phylogenomic research is accelerating the publication of landmark studies that aim to resolve deep divergences of major organismal groups . Meanwhile , systems for identifying and integrating the products of phylogenomic inference–such as newly supported clade concepts–have not kept pace . However , the ability to verbalize node concept congruence and conflict across multiple , in effect simultaneously endorsed phylogenomic hypotheses , is a prerequisite for building synthetic data environments for biological systematics and other domains impacted by these conflicting inferences . Here we develop a novel solution to the conflict verbalization challenge , based on a logic representation and reasoning approach that utilizes the language of Region Connection Calculus ( RCC–5 ) to produce consistent alignments of node concepts endorsed by incongruent phylogenomic studies . The approach employs clade concept labels to individuate concepts used by each source , even if these carry identical names . Indirect RCC–5 modeling of intensional ( property-based ) node concept definitions , facilitated by the local relaxation of coverage constraints , allows parent concepts to attain congruence in spite of their differentially sampled children . To demonstrate the feasibility of this approach , we align two recent phylogenomic reconstructions of higher-level avian groups that entail strong conflict in the "neoavian explosion" region . According to our representations , this conflict is constituted by 26 instances of input "whole concept" overlap . These instances are further resolvable in the output labeling schemes and visualizations as "split concepts" , which provide the labels and relations needed to build truly synthetic phylogenomic data environments . Because the RCC–5 alignments fundamentally reflect the trained , logic-enabled judgments of systematic experts , future designs for such environments need to promote a culture where experts routinely assess the intensionalities of node concepts published by our peers–even and especially when we are not in agreement with each other .
Three years ago , Jarvis et al . ( 2014; henceforth 2014 . JEA ) [1] published a landmark reconstruction of higher-level bird relationships . Within 12 months , however , another analysis by Prum et al . ( 2015; henceforth 2015 . PEA ) [2] failed to support several of the deep divergences recovered in the preceding study , particularly within the Neoaves sec . ( secundum = according to ) Sibley et al . ( 1988 ) [3] . Thomas ( 2015 ) [4] used the term "neoavian explosion" to characterize the lack of congruence between inferences of early-diverging lineages ( see also [5] ) . Similarly , after reviewing six phylogenomic studies , Suh [6] concluded that the root region of the Neoaves constitutes a "hard polytomy" . Multiple analyses have dissected the impact of differential biases in terminal and genome sampling , as well as evolutionary modeling and analysis constraints , on resolving this complex radiation [7 , 8 , 9] . Suh [6] argues that a well resolved consensus is not imminent ( though see [10] ) . Brown et al . ( 2017 ) [11] analyzed nearly 300 avian phylogenies , finding that the most recent studies "continue to contribute new edges" . These recent advancements provide an opportunity to reflect on how synthesis should be realized in the age of phylogenomics [11 , 12 , 13] . The neoavian explosion can be considered a use case where multiple studies provide strong signals for conflicting hierarchies . Resolution towards a single , universally adopted tree is unlikely in the short term . Rather than focusing on the analytical challenges along the path towards unitary resolution [9] , we turn to the issue of how the persistence of conflict affects the design of synthetic data infrastructures . In other words , how do we build a data service for phylogenomic knowledge in the face of persistent conflict ? This question is of broad relevance to systematists , comparative evolutionary biologists , and designers of biological information services interested in robust , reproducible , and reusable phylogenomic data . And it turns on the issue of improving identifiers and identifier-to-identifier relationships for this domain . Particularly verbal representations of the neoavian explosion are not well designed for conflict representation and synthesis [14] . To alleviate this , some authors use tree alignment graphs in combination with color and width variations to identify regions ( edges ) of phylogenomic congruence and conflict [15] . Other authors may show multiple incongruent trees side-by-side , using color schemes for congruent clade sections [9] . Yet others may use tanglegrams that are enhanced to highlight congruence [4] , rooted galled networks [16] or neighbor-net visualizations [17] that show split networks for conflicting topology regions , or simply provide a consensus tree in which incongruent bifurcating branch inferences are collapsed into polytomy [6] . Verbalizing phylogenomic congruence and conflict in open , synthetic knowledge environments [13] constitutes a novel challenge for which traditional naming solutions in systematics are inadequate . The aforementioned studies implicitly support this claim . All use overlapping sets of Code-compliant [18] and other higher-level names in the Linnaean tradition , with sources including [19] or [20] . To identify these source-specific name usages , we will utilize the taxonomic concept label convention of [14] . Accordingly , name usages sec . 2014 . JEA are prefixed with "2014 . " , whereas name usages sec . 2015 . PEA are prefixed with "2015 . " We diagnose the verbalization challenge as follows . ( 1 ) In some instances , identical clade names are polysemic–i . e . , have multiple meanings–across studies . For instance , 2015 . Pelecaniformes excludes 2015 . Phalacrocoracidae , yet 2014 . Pelecaniformes includes 2014 . Phalacrocoracidae; reflecting on two incongruent meanings of "Pelecaniformes" . ( 2 ) In other cases , two or more non-identical names have congruent meanings , e . g . , 2015 . Strisores and 2014 . Caprimulgimorphae . ( 3 ) Names that are unique to just one study–e . g . , 2015 . Aequorlitornithes or 2014 . Cursorimorphae–are not always reconcilable in meaning without additional human effort , thereby adding an element of referential uncertainty to the apparent conflict . ( 4 ) Lastly , many of the newly inferred and conflicting edges are not named at all . There is an implicit preference for labeling edges when suitable names are already available . However , unnamed edges can create situations where conflict cannot be verbalized and reconciled in a data environment , due to the lack of syntactic structure ( "names" ) . Jointly , the effects of polysemic names , synonymous names , exclusive yet hard-to-reconcile names , and conflicting unnamed edges are symptomatic of an information culture that is not ready for the identifier and identifier-to-identifier relationship challenges inherent in representing phylogenomic conflict . Suppose we wish to build a collaborative knowledge environment towards inferring "the tree of life" ( though see [12] ) . The design should allow us to individually represent and at the same integrate conflicting hierarchies , from the tips to the root . The system should respond to name-based data queries across these hierarchies , and return whether they are congruent or how they conflict in meaning . Clearly , the name usages of each individual source are not suited for this integration task . Traditional , Linnaean conventions allow for names to have evolving phylogenomic meanings across hierarchies and are therefore too under-powered for our purpose [21] . At root , this is a novel conceptual challenge for systematics and comparative evolutionary biology , made imperative by the accelerated generation and ingestion of phylogenomic trees into open , dynamic knowledge bases for reliable integration and re-use [11 , 13 , 22 , 23 , 24] . The services that such environments aspire to provide require an appropriate theory of node identity , and hence a conception of multi-node congruence or incongruence across individual trees and entire synthesis versions . Here we propose a solution to the phylogenomic conflict representation challenge . This solution requires collaboration between systematic experts , platform designers , and users of phylogenomic information . It is an extension of prior "concept taxonomy" research [14 , 25 , 26] , and deploys logic reasoning to align tree hierarchies based on Region Connection Calculus ( RCC–5 ) assertions of node congruence [27 , 28 , 29] . We demonstrate the feasibility of this approach by aligning subregions and entire phylogenomic trees inferred by 2015 . PEA and 2014 . JEA . In doing so , we address key representation challenges; such as the paraphyly of classification schemes used to label tree regions , and the inference of higher-level node congruence in spite of differentially sampled terminals . The alignment products for this use case constitute a novel answer to our central question: "how to build a synthetic knowledge environment in the face of persistent phylogenomic conflict ? " The discussion focuses on the feasibility and desirability of creating such an integration service , emphasizing the role of trained expert judgment in providing them [30] .
1 . Taxa are models , concepts are mimics . We typically refrain from using the terms "taxon" , "taxa" , or "clade ( s ) " . We take taxa to constitute evolutionary , causally sustained entities whose members are manifested in the natural realm . The task for systematics is to successively approximate the identities and limits of these entities . Thus , we assign the status of 'models' to taxa , which systematists aim to 'mimic' through empirical theory making . This perspective allows for realism about taxa , and also for the possibility to let our representations stand for taxa [31] , at any given time and however imperfectly , to support evolutionary inferences . In reserving a model status for taxa , we can create a separate design space for the human theory- and language-making domain . In the latter , we speak only of taxonomic or phylogenomic concepts–the products of inference making [21] . 2 . Sameness is limited to the same source . Therefore , for the purpose of aligning the neoavian explosion use case , we need not speak of the "same taxa" or "same clades" at all . Similarly , we need not judge whether one reconstruction or the other more closely aligns with deep-branching avian taxa , i . e . , which is ( more ) 'right' ? Instead , our alignment is only concerned with modeling congruence and conflict across two sets of concept hierarchies . The concepts are labeled with the "sec . " convention to maintain a one-to-one modeling relationship between concept labels and concepts ( clade identity theories ) . Accordingly , there is also no need to say that , in recognizing each a concept with the taxonomic name Neornithes , the two author teams are authoring "the same concept" . Instead , we model the two labels 2015 . Neornithes and 2014 . Neornithes , each of which symbolizes an individually generated phylogenomic theory region . As an outcome of our alignment , we may say that these two concepts are congruent , or not , reflecting the intensional alignment ( to be specified below ) of two phylogenomic theories . But , by virtue of their differential sources ( authorship provenance ) , the two concepts 2015 . Neornithes and 2014 . Neornithes are never "the same" . "Sameness" is limited in our approach to concepts whose labels contain an identical taxonomic name and which originate from a single phylogenomic hierarchy and source . That is , 2015 . Neornithes and 2015 . Neornithes are ( labels for ) the same concept . The methods used herein are consistent with [14 , 26 , 32] . They utilize three core conventions: ( 1 ) taxonomic concept labels to identify concepts; ( 2 ) is_a relationships to assemble single-source hierarchies via parent/child relationships; and ( 3 ) RCC–5 articulations to express the relative congruence of concept regions across multi-sourced hierarchies . The RCC–5 articulation vocabulary entails ( with corresponding symbol ) : congruence ( = = ) , proper inclusion ( > ) , inverse proper inclusion ( < ) , overlap ( >< ) , and exclusion ( ! ) . Disjunctions of these articulations are a means to express uncertainty; as in: 2015 . Neornithes {= = or > or <} 2014 . Neornithes . All possible disjunctions generate a lattice of 32 relationships ( R32 ) , where the "base five" are the most logically constraining subset [33] . The alignments are generated with the open source Euler/X software toolkit [28] . The toolkit ingests multiple trees ( T1 , T2 , T3 , etc . ) and articulation sets ( A1–2 , A2–3 , etc . ) , converting them into a set of logic constraints . Together with other default or facultative constraints ( C ) needed for modeling tree hierarchies , these input constraints are then submitted to a logic reasoner that provide two main services . First , the reasoner infers whether all input constraints are jointly logically consistent , i . e . , whether they permit at least one "possible world" . Second , if consistency is attained , the reasoner infers the set of Maximally Informative Relations ( MIR ) . The MIR constitute that unique set of RCC–5 articulations for every possible concept pair across the input sources from which the truth or falseness of any relationship in the R32 lattice can be deduced [14 , 26 , 33] . Many toolkit options and functions are designed to encode variable alignment input and output conditions , and to interactively obtain adequately constrained alignments . The toolkit also features a stylesheet-driven alignment input/output visualization service that utilizes directed acyclical graphs [28] . A step-wise account of the user/toolkit workflow interaction is provided in [26] . Aligning phylogenomic trees entails several special representation and reasoning challenges . We address three aspects here that have not been dealt with extensively in previous publications . 1 . Representing intensional parent concept congruence via locally relaxed coverage . The first challenge relates directly to the issue of parent node identity . Unlike comprehensive classifications or revisions [14 , 26 , 34] , phylogenomic reconstructions typically do not aspire to sample low-level entities exhaustively . Instead , select exemplars are sampled among all possible low-level entities . The aim is to represent lower-lever diversity sufficiently well to infer reliable higher-level relationships . Often , terminal sampling is not only incomplete for any single reconstruction , but purposefully complementary to that of other analyses . Generating informative genome-level data remains resource-intensive [10] . This makes it prudent to coordinate terminal sampling globally , by prioritizing the reduction of gaps over redundant terminal sampling . In the case of 2015 . PEA ( 198 terminals ) versus 2014 . JEA ( 48 terminals ) , only 12 species-level concept pairs have labels with identical taxonomic names . By default , the logic toolkit applies a coverage constraint to every input concept region . Coverage means that the region of a parent is strictly circumscribed by the union of its children [35] . However , this constraint is relaxable , either globally for all concepts , or locally for select concepts . To relax coverage locally , the prefix "nc_" ( no coverage ) is used in the input , as in 2014 . nc_Psittacidae . This means: either a parent concept's referential extension is circumscribed by the union of its explicitly included children , or there is a possibility of additional children being subsumed under that parent but not mentioned in the source phylogeny . Either scenario can yield consistent alignments . In other words , if a parent concept has relaxed coverage , it can attain congruence with another parent concept in spite of each parent having incongruent sets of child concepts . Managing coverage in the toolkit input is not trivial . Relaxing coverage globally is akin to saying "anything goes" , i . e . , any parent could potentially include any child . This would yield innumerable possible worlds , and therefore has no value for our purpose . On the other hand , applying coverage globally means–counter-intuitively in the case of phylogenomic trees–that only parents with completely congruent sets of children can themselves attain congruence . The challenge for experts providing the input is thus to relax coverage locally , and strictly in the service of 'neutralizing' lower-level sampling differences between trees that should not yield conflict at higher levels . The effect of locally relaxed coverage is illustrated in Figs 1–4 , using the example of parrots– 2015 . /2014 . Psittaciformes . At the species level , the author teams sampled wholly exclusive sets of concepts for this alignment region ( Figs 1 and 3 ) . Even at the genus level , only 2015 . /2014 . Nestor is redundantly sampled , yet with the articulation: 2015 . Nestor_meridionalis ! 2014 . Nestor_notabilis at the child level . Therefore , if no species-level concept sec . 2015 . PEA has an explicitly sampled and congruent region in 2014 . Psittaciformes , and , vice-versa , no species-level concept sec . 2014 . JEA has such a region in 2015 . Psittaciformes , then under global application of the coverage constraint we obtain the alignment: 2015 . Psittaciformes ! 2014 . Psittaciformes ( Fig 2 ) . The absence of even partial concept region overlap at the terminal level 'propagates up' to the highest-level parent concepts , which are therefore also exclusive of each other . Asserting higher-level node congruence in light of lower-level node incongruence requires a conception of node identity that affirms counter-factual statements of the following type: if 2014 . JEA had sampled 2014 . Psittacus_erithacus , then the authors would have included this species-level concept as a child of 2014 . Psittacidae . This is to say that 2015 . /2014 . Psittacidae , and hence their respective parents , are intensionally defined [25 , 36 , 37] . Using a combination of published topological information ( and support ) , more or less direct reiterations of phenotypic traits ( cf . discussions and supplementary data of 2015 . PEA and 2014 . JEA ) , and trained judgment [30] , we align these concept regions as if there are congruent property criteria that each region entails , i . e . , something akin to an implicit set of synapomorphies or uniquely diagnostic features . Of course , the phylogenomic data provided by 2015 . PEA and 2014 . JEA do not signal intensional definitions directly . But neither do their genome-based topologies for parrots provide evidence to challenge the status of such definitions as previously proposed [38] . In addition , particularly 2015 . PEA ( supplementary information; sections on "detailed justification for fossil calibrations" and "detailed phylogenetic discussion; pp . 3–21 ) provide a provide an in-depth account of how their preferred topology relates to published , property-centered circumscriptions of dozens of higher-level clade concepts . We have to assume , fallibly and non-trivially , that such topology-to-synapomorphy relations are also implied by JEA . 2014 , as reflected ( inter alia ) in their discussion . Three clarifications are in order . First , Region Connection Calculus is at best a means of translating the signal of an intensional definition . The congruent ( = = ) symbol means , only: two regions are congruent in their extension . The RCC–5 vocabulary is obviously not appropriate for reasoning directly over genomic or phenomic property statements . The reasoner does not assess whether 2015 . Psittacidae , or any included child or aligned concept , has 'the relevant synapomorphies' . Doing so would not be trivial even if property-based definitions were provided for all higher-level node concepts , because we would still have to make theory-laden assumptions about their congruent phylogenomic scopes [26 , 39 , 40] . Second , we are not providing detailed textual narratives that would justify each assertion of higher-level congruence . Such narratives are possible , and even needed to understand disagreements , because they explain the reasoning process behind an expert-made assertion . However , our main objective here is to focus on the issue of RCC–5 translation of systematic signals; not on a character-by-character dissection of each congruent articulation . Third , a sensible intensional alignment strategy uses a minimal number of instances of locally relaxed coverage in order to compensate for differential child sampling at lower levels , so that parent coverage can remain in place at higher levels to expose incongruent node concepts . The benefits of this strategy will be shown below . 2 . Representing clade concept labels . Our modeling approach requires that every region in each source tree receives a taxonomic or clade concept label . However , the source publications only provide such labels for a subset of the inferred nodes . In particular , 2015 . PEA ( p . 570: Fig 1 ) obtained 41 nodes above the ordinal level . Of these , 17 nodes ( 41 . 5% ) were explicitly labeled in either the published figure or supplement ( pp . 9–12 ) . The authors also cite [20] as the primary source for valid name usages , yet that list is not concerned with supra-ordinal names . Similarly , 2014 . JEA ( p . 1322: Fig 1 ) inferred 37 nodes above the ordinal level , of which 23 nodes ( 62 . 2% ) were given an explicit label . They provide an account ( cf . supplementary materials SM6: 22–24 ) of their preferred name usages , sourced mainly to [20] and [41] . In assigning clade concept labels at the supra-ordinal level when the authors may have failed to do so ( consistently ) , we nevertheless made a good faith effort–through examination of the supplementary information and additional sources [1 , 3 , 42 , 43 , 44 , 45 , 46 , 47]–to represent the authors' preferred name usages . Where usages were not explicit , we selected the only or most commonly applied clade concept name at the time of publication . This effort yielded 13 additional labels for 2015 . PEA ( Table 1 ) , and 7 such labels for 2014 . JEA ( Table 2 ) . If no suitable label was available , we chose a simple naming convention of adding "_Clade1" , "_Clade2" , etc . , to the available and immediately higher-level node label , e . g . 2014 . Passerea_Clade1 . The numbering of such labels along the tree topology starts with the most immediate child of a properly named parent , and typically follows down one section of the source tree entirely ( "depth-first" ) , before continuing with the higher-level sister section . Using this approach , we added 11 labels for 2015 . PEA ( Table 1 ) and 7 labels for JEA . 2014 ( Table 2 ) . If greater numbers of labels need to be generated , including siblings , then it is sensible to have a rule for ordering sibling nodes , e . g . by assigning the next-lowest number to the sibling whose child's name appears first in the alphabet . Our numbering of the labels 2014 . Passerea_Clade2 ( child with first-appearing letter: 2014 . Ardeae ) and 2014 . Passerea_Clade3 ( child: 2014 . Cursorimorphae ) adhere to this rule . The clade concept labeling convention was not applied below the family level , where instead phylogenomic resolution was collapsed into polytomy ( exception: Figs 1–4 ) . In the case of 2014 . JEA , only four family-level concepts include two children , whereas the remainder have a single child sampled . Resolving the monophyly of subfamilial clade concepts was not the primary aim of 2014 . JEA . The same applies to 2015 . PEA , who sampled 104/125 family-level concepts with only 1–2 children . 3 . Representing phylogeny/classification paraphyly . A third , relatively minor challenge is the occurrence of clade concepts in 2015 . PEA's phylogenomic tree that are not congruently aligned with higher-level concepts of [20] . We highlight these instances here because they represent a widespread phenomenon in phylogenomics . It is useful to understand how such discrepancies can be modeled with RCC–5 alignments ( Figs 5 and 6 ) . Fig 5 exemplifies the phylogenetic tree/classification incongruence observed in 2015 . PEA . The authors state ( supplementary Table 1 , p . 1 ) : "Taxonomy follows Gill and Donsker ( 2015; fifth ed ) " . As shown in Fig 5 , their phylogeny accommodates four sampled genus-level concepts that would correspond to children of the family-level concept Eurylaimidae sec . Gill & Donsker ( 2015 ) [20] . However , these concepts are arranged paraphyletically in relation to the reference classification . There is no parent concept that can be labeled 2015 . Eurylaimidae and would not also ( 1 ) include 2015 . Pittidae , i . e . , 2015 . Passeriformes_Clade1 in Fig 6 , or ( 2 ) just represent aligned subset of the Eurylaimidae sec . Gill and Donsker ( 2015 ) [20] , i . e . , 2015 . Passeriformes_Clade2 or 2015 . Passeriformes_Clade3 in Fig 6 . The concept Eurylaimidae sec . Gill and Donsker ( 2015 ) [20] has an overlapping ( >< ) articulation with 2015 . Passeriformes_Clade1 . In summary , our approach represents non-monophyly as an incongruent alignment of the phylogenomic tree and the source classification used to provide labels for that tree's monophyletic clade concepts . There are four distinct regions in the phylogeny of 2015 . PEA where such alignments are needed: {Caprimulgiformes , Eurylaimidae , Hydrobatidae , Procellariidae , Tityridae} sec . Gill & Donsker ( 2015 ) [20] . Each of these is provided in the S7–S9 Files . The source phylogenies specify 703 and 216 clade or taxonomic concepts , respectively . The frequent instances of locally relaxed coverage increase the reasoning complexity in relation to multi-classification alignments [14] , making specialized RCC–5 reasoning useful [48] . The reasoning and visualization challenges commend a partitioned alignment approach . To keep the Results concise , we show visualizations of the larger input and alignment partitions only in the Supporting Information . A detailed account of the input configuration and partitioning workflow is given below . Underlying all alignments is the presumption that at the terminal ( species ) level , the taxonomic concept labels of 2015 . PEA and 2014 . JEA are reliable indicators of either pairwise congruence or exclusion [14 , 26 , 32] . That is , e . g . , 2015 . Cariama_cristata = = 2014 . Cariama_cristata , or 2015 . Charadrius_hiaticula ! 2014 . Charadrius_vociferus . Because the time interval separating the two publications is short in comparison to the time needed for taxonomic revisions to effect changes in classificatory practice , the genus- or species-level taxonomic concepts are unlikely to show much incongruence; though see [49] or [50] . We note that 2015 . PEA ( p . 571 ) use the label 2015 . Urocolius ( _indicus ) in their phylogenomic tree , which also corresponds to the genus-level name endorsed in [20] Gill & Donsker ( 2015 ) . However , in their Supplementary Table 1 the authors use 2015 . Colius_indicus . We chose 2015 . Urocolius and 2015 . Urocolius_indicus as the labels to apply in the alignments . The toolkit workflow favors a partitioned , bottom-up approach [29] . The process of generating , checking , and regenerating input files must be handled 'manually' on the desktop ( note: improved workflow documentation and semi-automation of input-output-input changes are highly desirable ) . The performance of different toolkit reasoners was benchmarked in [28] . To work efficiently , the large problem of aligning all concepts at once is broken down into multiple smaller alignment problems , e . g . 2015 . /2014 . Psittaciformes ( Figs 3 and 4 ) . To manage one particular order-level alignment , we start with assembling each input phylogeny separately , with relaxed coverage applied as needed ( Fig 3 ) . The RCC–5 articulations for low-level concept pairs are provided incrementally , e . g . , in sets of 1–5 articulations at a time . Following such an increment , the toolkit reasoning process is re-/deployed to validate input consistency and infer the number of possible worlds . There is an option to specify that only one possible world is sought as output , which is equivalent to just checking for input consistency , as opposed to inferring all possible worlds . Doing so saves time as long as the input remains ( vastly ) under-specified . The stepwise approach of adding a small number of articulations at a time leads to increasingly constrained alignments , while minimizing the risk of introducing many new . difficult-to-diagnose inconsistencies . Once a set of small , topographically adjacent alignment partitions is well specified , these can serve as building blocks for the next , larger partition . Hence , the basic sequence of building up larger alignments is: ( 1 ) obtain a well-specified low- ( order- or family- ) level alignment; ( 2 ) record the inferred parent-level articulations from this alignment; ( 3 ) propagate the latter–now as low-level input articulations–for the next , more inclusive alignment; ( 4 ) as needed , prune the lowest-level ( sub-ordinal ) input concepts and articulations of ( 1 ) from this alignment; ( 5 ) repeat ( 1 ) to ( 4 ) for another paired region; ( 6 ) assemble the more inclusive alignment by ( manually ) connecting the pruned , propagated concepts and articulations from two or more lower-level alignments , by adding to them the higher-level concepts from each input phylogeny . Depending on the interplay between ( ranked ) higher-level names recognized in each phylogeny and the number of terminal concepts sampled , steps ( 1 ) to ( 6 ) may be iterated once ( e . g . , 2015 . /2014 . {Falconiformes , Psittaciformes} ) or multiple times ( e . g . , 2015 . /2014 . Passeriformes ) to cover a supra-/ordinal alignment . An example of the latter is the 2015 . /2014 . Passerimorphae alignment , which includes two order-level concepts and their children in each source phylogeny . Such mid-level partitions eventually form the basis for the largest alignment partitions , e . g . 2015 . /2014 . Telluraves . Sometimes , coverage will have to be relaxed even at higher levels . In all , 2014 . JEA sample children of 34 order-level concepts in their phylogeny , whereas 2015 . PEA recognize 40 order-level concepts . The latter authors represent four order-level concepts for which no analogous children are included in 2014 . JEA , i . e . : 2015 . {Apterygiformes , Casuariiformes , Ciconiiformes , Rheiformes} . Three of these are assigned to 2015 . Palaeognathae , whereas 2015 . Ciconiiformes are subsumed under 2015 . Pelecanimorphae–in each case under relaxed parent coverage . The remaining 36 order-level concepts sec . 2015 . PEA show some child-level overlap with those of 2014 . JEA . Our partitioning approach for this use case started with specifying the input constraints for nearly 35 paired order-level concepts and their respective children , as demonstrated in Figs 3 and 4 . The largest order-level partition is 2015 . /2014 . Passeriformes , with 148 x 22 input concepts , seven instances of relaxed parent coverage , and 101 input articulations . This alignment completes in less than 15 seconds on an individual 2 . 0 GHz processor , yielding 3 , 256 MIR . As the partitions grew , we configured the following six , non-overlapping alignments as building blocks for the global alignment: 2015 . /2014 . Palaeognathae ( 34 x 12 input concepts , four instances of relaxed coverage , and 25 articulations; same data sequence used for following alignments ) , 2015 . /2014 . Galloanserae ( 49 , 16 , 7 , 46 ) , 2015 . Columbaves/2014 . Columbimorphae + 2014 . Otidimorphae ( 53 , 37 , 13 , 37 ) , 2015 . Strisores/2014 . Caprimulgimorphae ( 44 , 17 , 8 , 32 ) , 2015 . /2014 . Ardeae ( 100 , 55 , 19 , 75 ) , and the largest partition of 2015 . /2014 . Telluraves ( 316 , 104 , 37 , 241 ) . At the next more inclusive level , the inferred congruence of 2015 . Telluraves = = 2014 . Telluraves presented an opportunity to partition the entire alignment into two similarly sized regions , where the complementary region includes all 2015 . /2014 . Neornithes concepts ( 392 , 174 , 58 , 259 ) , except those subsumed under 2015 . /2014 . Telluraves , which are therein only represented with two concepts labels and one congruent articulation . These two complements are the core partitions that inform our use case alignment , globally . The corresponding S10 and S11 Files include the input constraint ( . txt ) and visualization ( . pdf ) files , along with the alignment visualization ( . pdf ) and MIR ( . csv ) . The two large partitions yield unambiguous RCC–5 articulations from the species concept level to that of 2015 . /2014 . Neornithes . They can be aggregated into a synthetic , root-to-order level alignment , where all subordinal concepts and articulations are secondarily pruned away ( see above ) . Such an alignment retains the logic signal derived from the bottom-up approach , but represents only congruent order-level concept labels as terminal regions , except in cases where there is incongruence . We present this alignment as an analogue to Fig 1 in [4] ( p . 515 ) , and compare how each conveys information about congruent and conflicting higher-level clade concepts . Lastly , we further reduce the root-to-order alignment to display only 5–6 clade concept levels below the congruent 2015 . /2014 . Neoaves . This region of the alignment is the most conflicting , and therefore forms the basis for our Discussion .
Our alignments show widespread higher-level congruence across the neoavian explosion use case; along with several minor regions of conflict and one strongly conflicting region between concepts placed immediately below the 2015 . /2014 . Neoaves . We focus first on the large complementary partitions , i . e . 2015 . /2014 . Neornithes ( without ) / 2015 . /2014 . Telluraves ( see S10 and S11 Files ) . Jointly , they entail 707 concepts sec . 2015 . PEA and 283 concepts sec . 2014 . JEA . Among these , 34 "no coverage" regions were added to 2015 . PEA's phylogeny , whereas 61 instances of relaxing parent coverage were assigned to 2014 . JEA's phylogeny , for a total of 95 instances of relaxing this constraint . The 2015 . /2014 . Neornithes partition shows 305 aligned regions– 247 without the "no coverage" regions–of which 60 congruently carry at least one concept label from each source phylogeny . This alignment also shows eight congruent species-level concept regions . These would be the only instances of congruence if coverage were globally applied ( Figs 1 and 2 ) . Therefore , relaxing the coverage constraint yields 52 additional instances of higher-level node congruence . Similarly , the 2015 . /2014 . Telluraves partition has 231 aligned regions– 194 without the "no coverage" regions–of which 38 are congruent . This corresponds to an increase of 34 regions , compared to four congruent species-level concept regions present under strict coverage . Correcting for the redundant 2015 . /2014 . Telluraves region , we 'gain' 85 congruent parent node regions across the two phylogenies if node identity is encoded intensionally ( Figs 3 and 4 ) . Indeed , this approach yields the intuitive articulation 2015 . Neornithes = = 2014 . Neornithes at the highest level . We now focus on characterizing the conflict between 2015 . PEA and 2014 . JEA . Phylogenomic incongruence can be divided into two general categories: ( 1 ) differential granularity or resolution of clade concepts ( RCC–5 translation: < or > ) , and ( 2 ) overlapping clade concepts ( RCC–5 translation: >< ) . The first of these is less problematic from a standpoint of achieving integration: for a given alignment subregion , the more densely sampled phylogeny will entail additional , more finely resolved clade concepts in comparison to its counterpart . Typically , this distinction belongs to the phylogeny of 2015 . PEA , due to the 4:1 ratio of terminals sampled . There are 83 above species-level clade concepts sec . 2015 . PEA that can be interpreted as congruent refinements of the 2014 . JEA topology ( see S10 and S11 Files ) . Conversely , only two such instances of added resolution are contributed by 2014 . JEA: ( 1 ) 2014 . Passeriformes_Clade3 which entails 2014 . Passeridae and 2014 . Thraupidae; and ( 2 ) 2014 . Haliaeetus with two subsumed species-level concepts . Nevertheless , the joint 97 congruent node regions and 85 refining node regions cover a large section of the alignment where integration is either reciprocally ( = = ) or unilaterally ( < or > ) feasible . The remaining 38 instances of overlapping articulations between constitute the most profound conflict . These instances are clustered in four distinct regions , i . e . : 2015 . /2014 . Pelecanimorphae ( 8 overlaps; Fig 7 and S12 File ) ; 2015 . Passeri/2014 . Passeriformes_Clade2 ( 3 overlaps; Fig 8 and S13 File ) ; 2015 . Eutelluraves/2014 . Afroaves ( 1 overlap; Figs 9 and 10 , and S14 and S15 Files ) ; and finally , 2015 . /2014 . Neoaves ( 26 overlaps; Figs 11–13 , and S16–S18 Files ) . We will examine each of these in sequence . 1 . 2015 . /2014 . Pelecanimorphae . The two author teams sampled four family-level concepts congruently for this alignment region ( Fig 7 ) . However , 2015 . PEA's phylogeny entails six additional family-level concepts that have no apparent match in 2014 . JEA . Moreover , the latter authors recognize only one order-level concept , 2014 . Pelecaniformes , under which all four family-level concepts are subsumed , including 2014 . Phalacrocoracidae . In contrast , 2015 . PEA infer an intensionally less inclusive concept of 2015 . Pelecaniformes , and place their congruent 2015 . Phalacrocoracidae in the order-level concept 2015 . Suliformes . This is the first instance where we may plausibly reject the proposition: "Had 2014 . JEA sampled 2014 . Phalacrocoracidae , they would have assigned this concept to 2014 . Suliformes" . The assertion is no longer counter-factual: 2014 . JEA did sample the corresponding child concept ( 2014 . Phalacrocoracidae ) , but did not assign it to a parent concept separate from 2014 . Pelecaniformes . Accordingly , we obtain three overlapping , 'cascading' articulations between concepts that form the 2015 . Suliformes higher-level topology and 2014 . Pelecaniformes . Meanwhile , the uniquely sampled 2015 . Ciconiiformes are subsumed under 2014 . Pelecanimorphae with relaxed parent coverage . Within 2015 . Pelecaniformes , we obtain five additional overlapping articulations between five concepts that make up the 2015/2014 supra-familial topologies in this alignment ( Fig 7 ) . This conflict is due to the differential assignment of 2015 . /2014 . Pelecanidae . Specifically , 2015 . PEA inferred a sister relationship of 2015 . Pelecanidae with 2015 . Balaenicipitidae , for which 2014 . JEA have no sampled match . Meanwhile , the latter authors inferred a sister relationship of 2014 . Pelecanidae with 2014 . Ardeidae . The latter concept is matched in 2015 . PEA with 2015 . Ardeidae , though not as the most immediate sister concept of 2015 . Pelecanidae . Of course , we may posit that a 2015 . Ardeidae/2015 . Pelecanidae sister relationship is what 2015 . PEA would have obtained , had these authors not also sampled 2015 . Balaenicipitidae and 2015 . Scopidae . But they did , and hence obtained two clade concepts that include 2015 . Pelecanidae yet exclude 2015 . Ardeidae; i . e . , 2015 . Pelecanoidea_Clade1 and 2015 . Pelecanoidea_Clade2 . While relaxing parent coverage for 2014 . Pelecaniformes_Clade2 could serve to mitigate this conflict , we deem the overlapping relationship to better represent 2015 . PEA's phylogenomic signal , which happens to 'break up' the lowest supra-familiar clade concept supported by 2014 . PEA . 2 . 2015 . Passeri/2014 . Passeriformes_Clade2 . This alignment region is another instance where relaxing parent coverage can only partially mitigate conflict ( Fig 8 ) . In this case , 2015 . PEA and 2014 . JEA sampled two sets of family-level concepts that are wholly exclusive of each other , except for 2015 . /2014 . Corvidae . Regarding the only two additional family-level concepts recognized in 2014 . JEA–i . e . , 2014 . Passeridae and 2014 . Thraupidae–we may posit counter-factually that these would be subsumed under 2015 . Passeroidea with relaxed coverage [47] . However , further assertions of congruence are difficult to justify , given the limited sampling of 2014 . JEA . Thus , in our current representation , 2014 . Passeriformes_Clade2 shows an overlapping relationship with 2015 . Passeroidea , its immediate parent 2015 . Passerida , and also with 2015 . Corvoidea . 3 . 2015 . Eutelluraves/2014 . Afroaves . A single overlap occurs just within the congruent parent concepts 2015 . /2014 . Telluraves ( Fig 9 ) . Two levels below this paired parent region , both author teams recognize three congruent children; viz . 2015 . /2014 . {Australaves , Coracornithia , Accipitrimorphae/Accipitriformes} . However , 2015 . Prum group the former two concepts under 2015 . Eutelluraves , with 2015 . Accipitriformes as sister; whereas 2014 . JEA cluster the latter two concepts under 2014 . Afroaves , with 2014 . Australaves as sister . This is the first occurrence of conflict that cannot justifiably be resolved by relaxing parent coverage , but instead reflects divergent phylogenomic signals . How to speak of such overlap ? In Fig 9 , we utilize clade concept labels that pertain to each input phylogeny . In the resulting alignment , the articulation 2015 . Eutelluraves >< 2014 . Afroaves is visualized as a dashed blue line between these regions . Yet Fig 9 also specifies the extent of regional overlap at the next lower level . Accordingly , only the region 2015 . /2014 . Coracornithia is subsumed under each of the overlapping parents . This is indicated by the two inclusion arrows that extend 'upward' from this region . The other two paired child regions are respectively members of one parent region . If we call the input regions 2015 . Eutelluraves "A" and 2014 . Afroaves "B" , we can use the following syntax to identify output regions that result from overlapping input concepts [26]: A*B ( read: "A and B" ) constitutes the output region shared by two parents , whereas A\b ( "A , not b" ) and B\a ( "B , not a" ) are output regions with only one parent . We call this more granular syntax split-concept resolution ( "merge concepts" in [26] ) , as opposed to whole-concept resolution which preserves the syntax and granularity provided by the input concept labels . In Fig 10 , the 2015 . /2014 . Telluraves overlap is represented with split-concept resolution . This eliminates the need to visualize a dashed blue line between 2015 . Eutelluraves and 2014 . Afroaves ( Fig 9 ) . Moreover , in this case the split-concept resolution syntax is redundant or unnecessary , because each of the three resolved regions under "A" ( 2015 . Eutelluraves ) and "B" ( 2014 . Afroaves ) is congruent with two regions already labeled in the corresponding input phylogenies . We will see , however , that this granular syntax is essential for verbalizing the outcomes of more complex alignments that contain many overlapping regions . 4 . 2015 . /2014 . Neoaves . The remaining 26 instances of overlap are shown under different alignment visualizations in Figs 11–13 . They occur 1–5 levels below the congruent concept pair 2015 . /2014 . Neoaves , and jointly make up the primary region of conflict between these reconstructions . Because parent coverage was already and selectively applied at lower levels , none of the 26 overlaps in the alignment are caused by differential child sampling . Therefore parent coverage must hold here , resulting in genuine conflict in the higher-level arrangement of congruent sets of children . Our Fig 11 is intended to be an RCC–5 alignment analogue to Fig 1 in [3] . The alignment reaches from the root to the ordinal level , and to the family level in the two subregions where order-level concepts are conflicting ( see Fig 4 and S12 File ) . The visualization provides an intuitive signal of the distribution of in-/congruence throughout the alignment . In all , 66/111 regions ( 59 . 5% ) are congruent , of which 22 are located in the 2015 . /2014 . Telluraves; 15 are contained in the 2015 . /2014 . Ardeae; and 5 are part of the 2015 . /2014 . Columbimorphae . Outside of the 2015 . /2014 . Neoaves , 8 such regions are present . In other words , the two phylogenies are congruent at the highest level and also in several intermediate regions above the ordinal level . Fig 12 shows just the neoavian explosion region under whole-concept resolution . Each phylogeny contributes 21 input concepts to this 'zoomed-in' alignment , which yields 13 congruent regions . Of these , only 2015 . /2014 . Neoaves and 2015 . /2014 . Otidimorphae represent non-terminal concepts . Unpacking the complexity of this conflict region requires a stepwise analysis . From the perspective of 2015 . PEA , the 2015 . Neoaves are split into a sequence of three unnamed , higher-level clade concepts , i . e . 2015 . {Neoaves_Clade1 , Neoaves_Clade2 , Neoaves_Clade3} , with 2015 . {Strisores , Columbaves , Gruiformes} as corresponding sister concepts . The two children of 2015 . Neoaves_Clade3 are 2015 . {Aequorlitornithes , Inopinaves} . The authors accept the nomenclature of [44] for 2015 . Strisores , with is congruent with 2014 . Caprimulgimorphae; and the region 2015 . /2014 . Gruiformes is congruent as well . However , the remaining six high-level concepts of 2015 . PEA are in conflict with the two highest-level neoavian concepts of 2014 . JEA , i . e . 2014 . {Columbea , Passerea} , and also with any of the four unnamed clade concepts below 2014 . Passerea . In particular , the node sequence 2015 . {Neoaves_Clade3 , Aequorlithornites , Aequorlithornites_Clade1} participates in 16/26 overlaps , as summarized in Table 3 . Loosely corresponding to this sequence are the concepts 2014 . {Passerea_Clade1 , Passerea_Clade2 , Cursorimorphae} , jointly with 10 overlaps . These overlaps are grounded in the incongruent assignment of five paired , lower-level concept regions; viz . 2015 . /2014 . {Ardeae , Charadriiformes , Opisthocomiformes , Phoenicopterimorphae , Telluraves} . Two conflicting placements contribute most to the number of overlaps: ( 1 ) 2015 . /2014 . Charadriiformes in 2015 . Aequorlithornites_Clade1 ( sister to 2015 . Phoenicopterimorphae ) versus 2014 . Cursorimorphae ( sister to 2014 . Gruiformes ) ; and ( 2 ) 2015 . /2014 . Phoenicopterimorphae in 2015 . Aequorlithornites_Clade1 versus 2014 . Columbea ( sister to 2014 . Columbimorphae ) . The newly proposed yet unnamed 2015 . Aequorlithornites_Clade1 , consisting of certain "waterbirds" , in effect causes the most topological incongruence with 2014 . JEA . This concept , together with its four superseding parents , 'triggers' 20/26 overlaps with the phylogenomic tree of 2014 . JEA . Two additional clusters of conflict are identifiable in Fig 12 . The first concerns the alignment of the two concepts 2015 . Inopinaves and 2014 . Passerea_Clade2 , which share the child regions 2015 . /2014 . Telluraves , yet which differentially accommodate the congruent regions 2015 . /2014 . Ardeae and 2015 . /2014 . Opisthocomiformes . This further contributes to the abundance of overlaps along the respective 2015 . Neoaves_Clade{1–3}/Aequorlithornites/_Clade1 and 2014 . Passerea/_Clade{1–3}/Cursorimorphae chains . Second , the two paired regions 2015 . /2014 . Columbimorphae and 2015 . /2014 . Otidimorphae are incongruently assigned to three overlapping parents , i . e . 2015 . Columbaves and 2014 . {Columbea , Passerea_Clade4} . From the perspective of 2015 . PEA , 2014 . JEA's bifurcation of 2014 . Columbea and 2014 . Passerea is the most conflicting , as these two concepts participate in 11 overlaps . A third and more minor incongruence concerns the placement of three concept regions within the 2015 . /2014 . Oditimorphae . In Fig 13 , the same 'zoomed-in' alignment is shown under split-concept resolution . This permits identifying all output regions created by the 26 overlaps of the neoavian explosion ( see Table 4 ) . The entire set consists of 78 labels; i . e . , 26 labels for each split-resolution product {A*B , A\b , B\a} for one instance of input region overlap . Not all of these split-concept resolution labels are semantically redundant with those provided in the input . Specifically , 51 labels are generated 'in addition' for the 12 terminal congruent regions ( compare with Fig 12 ) . These are indeed unnecessary synonyms for regions already identified in the input . However , the relative number of additional labels generated per input region is telling . This number will be highest for regions whose differential placements are the primary drivers of incongruence . As explained above , these are: 2015 . /2014 . {Phoenicopterimorphae , Charadriiformes , Columbimorphae} , respectively with 14 , 8 , and 7 additional labels . Six redundant split-concept resolution labels are further produced for input regions that are unique to one phylogeny; e . g . , 2014 . Columbea is also labeled 2015 . Neoaves_Clade1 \ 2014 . Passerea ( where the "\" means: not ) . The remaining 21 split-concept resolution labels identify 15 salmon-colored alignment regions– 11 uniquely and 4 redundantly with 2–3 labels each–for which there are no suitable labels in either of the phylogenomic input trees ( Table 4 ) . Forty-six additional articulations are inferred to align these regions to those displayed in Fig 12 . Although these novel regions are not congruent with any clade concepts recognized by the source phylogenies , they are needed to express how exactly the authors' respective clade concepts overlap . Three distinct reference services are gained by generating the split-concept resolution labels . First , in cases where no whole-concept resolution labels are available , we obtain appropriately short and consistent labels to identify the split regions caused by overlapping clade concepts . Second , the {A*B , A\b , B\a} triplets have an explanatory function , by using the same syntactic set of input labels ( A , B ) to divide complementary alignment subregions of an overlap . If we focus on one label of a triplet , we can find the two complements , and thereby systematically explore the 'reach' of each split in the alignment . Third , the clade concept labels ( A , B ) used in the split-concept resolution labels will be exactly those that identify overlapping regions across the source phylogenies . We can now also ask to what extent the clade names ( syntax ) used by the two author teams succeed or fail to identify congruent and incongruent concept regions ( semantics ) . Such name:meaning ( read: "name-to-meaning" ) analyses were carried out in three previous alignment use cases , with rather unfavorable outcomes for the respective names in use [14 , 32 , 51] . Here , based on the alignment of Fig 11 , the 97 x 83 input concepts yield a set of 8 , 051 MIR ( S16 File ) . Of these , 384 MIR involve one of four "no coverage" regions added to 2014 . JEA concepts . We therefore restrict the name:meaning analysis to the remaining 7 , 667 MIR ( Table 5 ) . Interestingly , the clades names used by the respective author teams fare rather well . Only nine of 7 , 667 pairings in the MIR ( 0 . 12% ) are unreliable as identifiers of in-/congruence of the respective RCC–5 articulation . In seven instances , two congruent concepts have different names . Four of these merely involve changes in name endings , viz . : 2015 . Accipitriformes = = 2014 . Accipitrimorphae , 2015 . Galloanserae = = 2014 . Galloanseres , 2015 . Gaviiformes = = 2014 . Gaviimorphae , and 2015 . Pteroclidiformes = = 2014 . Pterocliformes . The other three instances involve the respectively preferred roots 2015 . Strisores and 2014 . Caprimulgi-{formes , morphae} . The articulation 2015 . Pelecaniformes < 2014 . Pelecaniformes is the single instance in which the meaning of the same name is less inclusive in one source ( Fig 7 ) . Lastly , the overlapping relationship 2015 . Otidimorphae_Clade1 >< 2014 . Otidimorphae_Clade1 involves the same name ( Figs 12 and 13 ) , though it is not actually used by the author teams ( see Methods ) . In summary , the clade concept names used by 2015 . PEA and 2014 . JEA rarely provide an incorrect signal regarding in-/congruence . This desirable outcome seems to reflect their recognition that newly inferred clade concepts merit the use of unique names . We now compare these results with conflict analysis and visualization tools created for the Open Tree of Life project ( OToL ) –a community-curated tree synthesis platform [13 , 22 , 23 , 24] . The OToL approach is explained in [11 , 15 , 23 , 52 , 53] . The method starts off with 'normalizing' all terminal names in the source trees to a common taxonomy [24] . Having the same terminal name means taxonomic concept congruence ( = = ) . To assess conflict from the perspective of one rooted input tree ( A ) , a source edge j of that tree is taken to define a rooted bipartition S ( j ) = Sin | Sout , where Sin and Sout are the tip sets of the ingroup and outgroup , respectively . The algorithm progresses sectionally from the leaves to the root . Concordance or conflict for a given edge j in tree A with that of tree B is a function of the relative overlap of the corresponding tip sets , as follows [23] . Concordance between two edges in the input trees A and B is obtained when Bin is a proper subset ( ⊂ ) of Ain and Bout ⊂ Aout . On the other hand , two edges in trees A and B are conflicting if none of these sets are empty: Ain intersects ( ⋂ ) with Bin , Ain ⋂ Bout , or Bin ⋂ Aout . In other words , conflict means that there is reciprocal overlap in the ingroup and outgroup bipartitions across the two trees . We applied this approach in both directions , i . e . starting with 2014 . JEA as primary source and identifying edges therein that conflict with those of 2015 . PEA , and vice-versa . The visualizations are shown in Figs 14 and 15 , respectively . Most of the red edges in Fig 15 , which is based on the more densely sampled tree sec . 2015 . PEA , are consistent with the overlapping RCC–5 relationships shown in Figs 7 to 13 . However , within the 2015 . Pelicanimorphae , certain RCC–5 overlaps ( Fig 7 ) are not recovered ( "false positives" ) . In addition , numerous edges within the 2015 . Passeriformes are shown as conflicting ( "false negatives" ) but are congruent refinements based on the RCC–5 alignment ( Fig 8 ) . Using the less densely sampled tree sec . 2014 . JEA as the base topology creates is instructive ( Fig 14 ) . Here , a much larger subset of the topology 'backbone' is inferred by the OToL algorithm as conflicting–an outcome that would appear inconsistent . For instance , 2014 . {Neoaves , Ardeae , Coracornithia} are shown as conflicting edges in Fig 14 , when 2015 . {Neoaves , Ardeae , Coracornithia} are concordant edges in Fig 15 . The inconsistencies are caused by the addition of terminals sec . 2015 . PEA that have no matches in 2014 . JEA's sampled tips and tree , and will therefore attach as children to a higher-level parent in the OToL taxonomy . The latter is used to place terminals that are differentially sampled between sources . For instance , 2015 . Ciconiiformes–which has no close match in 2014 . JEA–may end up attaching as a child of 2014 . Neognathae instead of 2014 . Pelecanimorphae ( Fig 7 ) . Hence the OToL taxonomy is used to represent concept intensionality , but it cannot do so reliably if it lacks relevant input concepts . At the time of analysis , the OToL taxonomy lacked a name/concept for "Neoaves" . This means that the 2015 . /2014 . Neoaves ingroup/outgroup bipartitions will be inconsistent in evaluating the placement of 2015 . Ciconiiformes , showing conflict in Fig 14 but not in Fig 15 .
We review the key conventions of our approach before discussing services that can be derived from our alignments . What can we gain from this approach , both narrowly for this use case and for future data integration in systematics ? Data representation designs have inherent trade-offs . Unlike other semi-/automated phylogenomic conflict visualization methods [13 , 23 , 24] , the above approach requires extensive upfront application of human expertise to obtain the intended outcomes . In return , the RCC–5 alignments deliver a level of explicitness and verbal precision exceeding that of published alternatives [4 , 5 , 6 , 9 , 16 , 17] . We can not just verbalize all instances of congruence and conflict , but transparently document and therefore understand their provenance in a global alignment ( Figs 11 and 13 ) . In other words , the RCC–5 alignments provide a logically tractable means to identify and also explain the extent of conflict . We can derive novel data services from the alignment products . ( Note that these services are envisioned but not yet implemented in a web-based platform . ) Example queries include the following . ( 1 ) Show all congruent regions of the alignment and their clade concept labels . ( 2 ) Modify this query to only apply to alignment regions subsumed under one particular concept and source , such as 2014 . Columbea . ( 3 ) For any subset region of the global alignment ( e . g . , 2015 . /2014 . Australaves ) , show the lowest-level pairs of children that are sampled congruently , versus those that are sampled incongruently . ( 4 ) Highlight within such an alignment region all clade concepts for which parent coverage is relaxed , and which show congruence as a result of this action . ( 5 ) Highlight sets of concepts where incongruence is due to differential granularity ( sampling ) , versus actual overlap . ( 6 ) Identify and rank concepts that participate in the greatest number of overlapping relationships ( Table 3 ) . ( 7 ) Identify and rank the longest chains of nested , overlapping concept sets ( Fig 12 ) . ( 8 ) Highlight the congruent , lowest-level concept pairs whose incongruent placement into higher-level regions causes the chains of overlap . ( 9 ) List all split-concept resolution labels in complementary triplets {A*B , A\b , B\a} , and provide for each the two immediate children and ( again ) the set of lower-level , whole-concept resolution regions that are differentially distributed by the split ( Fig 13 and Table 4 ) . ( 10 ) Identify clade names that are unreliable across the source phylogenies; including identical clade name pairs that participate in concept labels with an incongruent relationship , or different clade names whose concept labels have a congruent relationship ( Table 5 ) . All of the above queries , and many others we could propose , are enabled by our RCC–5 representation and reasoning conventions , which therefore present a new foundation for building logic-based , machine-scalable data integration services for the age of phylogenomics . Conceptualizing node identity and congruence this way addresses a gap in current systematic theory that is not adequately filled by other syntactic solutions . Linnaean naming . We have shown elsewhere that homonymy and synonymy relationships are unreliable indicators of congruence [14 , 26 , 32] . Code-enforced Linnaean naming is designed to fixate the meaning of names by ostension , while allowing the intensional components to remain ambiguous [21 , 54 , 55 , 56 , 57] . This trade-off effectively shifts the burden of disambiguating varying intensionalities associated with Linnaean names onto an additional , interpreting agent–typically human experts . Our RCC–5 alignment approach can be viewed as a way to formalize the disambiguation effort , so that it can attain machine-interpretability . Phyloreferencing . Similarly , node-based phyloreferences [58 , 59 , 60] are not well suited to reconstruct an alignment such as that of 2015 . /2014 . Pelecanimorphae ( Fig 7 ) . This would require: ( 1 ) an elaborate notion of phyloreference homonymy and synonymy ( e . g . , 2015 . Pelecanifores versus 2014 . Pelecaniformes , or 2015 . Strisores versus 2014 . Caprimulgimorphae ) ; ( 2 ) node-based definitions with inclusion/exclusion constraints that cover all terminals in the phylogeny; and ( 3 ) synapomorphy-based definitions at higher levels to model the local relaxation of coverage constraints . All of these functions may be feasible in principle with phyloreferences , provided that human experts are permitted to enact them . However , it may be fair to say that phyloreferences were not mainly designed to bring out fine differences between node concepts across multiple phylogenies . They are best utilized when concept evolution and conflict are not the main drivers of an information system design . The two largest alignments of 2015 . /2014 . Neornithes ( without ) / 2015 . /2014 . Telluraves jointly entail 895 concepts and 95 instances of relaxed parent coverage . They provide us with 97 congruent regions in the global alignment , of which 85 regions are obtained only because of the indirect modeling of intensional node definitions . The contingency of the alignment outcome on expert intentions is neither surprising nor trivial . We should therefore explore this dependency more deeply . Redelings and Holder [23: pp . 5–6] comment on the OToL synthesis method: "Any approach to supertree construction must deal with the need to adjudicate between conflicting input trees . We choose to deal with conflict by ranking the input trees , and preferring to include edges from higher-ranked trees . The merits of using tree ranking are questionable because the system does not mediate conflicts based on the relative amount of evidence for each alternative . […] In order to produce a comprehensive supertree , we also require a rooted taxonomy tree in addition to the ranked list of rooted input trees . Unlike other input trees , the taxonomy tree is required to contain all taxa , and thus has the maximal leaf set . We make the taxonomy tree the lowest ranked tree . […] Our method must resolve conflicts in order to construct a single supertree . However , the rank information used to resolve conflicts is an input to the method , not an output from the method . We thus perform curation-based conflict resolution , not inference-based conflict resolution . " Clearly , the outcomes of the OToL synthesis method are also deeply dependent on expert input regarding the relative ranking of input phylogenies and of the OToL taxonomy [24] . We have shown ( Figs 14 and 15 ) that these choices can lead to inconsistent outcomes whenever the sequence of input trees determines how concordance and conflict are negotiated by the algorithms . If the less densely sampled tree is prioritized , and the taxonomy cannot accommodate all components of a lower-ranked tree , then the method will show more conflict in comparison to an inverse input sequence . Any global rule of priority among trees is a poor proxy for modeling individual node concept intensionality , which requires making reliable , local decisions between ( 1 ) conflict due to differential granularity versus ( 2 ) conflict due to overlap . We can now return to the challenge posed in the Introduction . How do we build a data service for phylogenomic knowledge in the face of persistent conflict ? Our answer is novel in the following sense . Assuming that such a service is desirable , we show that achieving it fundamentally depends on making and expressing upfront empirical commitments about the intensionalities of clade concepts whose children are incongruently sampled . Without embedding these judgments into the alignment input , we lose the 85 congruent parent regions recovered under relaxed parent coverage . We furthermore lose the ability to distinguish the former from more than 340 alignment regions that are not congruent . And we lose the power to express the nature of this residual conflict–granularity versus overlaps–and how to resolve it . In other words , the first step for building the phylogenomic data knowledge service will be to recognize that conceptualizations of node identity within such a system just cannot be provided through some mechanical , 'objective' criterion . Instead , we need an inclusive standard of objectivity that embraces trained judgment as an integral part of identifying and linking node concepts [30] . In that sense , phylogenomic syntheses are inference-based ( contra [23] ) and also driven by a specific purpose . As integrative biologists , our goal in providing RCC–5 alignments is to maximize intensional node congruence . There may not be a more reliable criterion for achieving this than expert judgment , which draws on complex and context-specific theoretical knowledge [40 , 43 , 61] . Logic representation and reasoning can help render these constraints explicit and consistent , and expose implicit articulations through the MIR which encompass all node concepts in an alignment . But logic cannot substitute the expert aligners' intensional aims and definitions . Building a phylogenomic data knowledge service forces us to become experts about externally generated results that conflict with those which we may ( currently ) publish or endorse . We need to become experts of another author team's node concepts , to the point where we are comfortable with expressing counter-factual statements regarding their intensionalities , in spite of incongruent child sampling . This will require a profound but necessary adjustment in achieving a culture of synthesis in systematics that no longer manages conflict this way: "If we do not agree , then it is either our view over yours , or we just collapse all conflicting node concepts into polytomies" . In contrast , we need to develop the following culture of synthesis: "We may not agree with you , but we understand your phylogenomic inference well enough to express our dis-/agreements in a logic-compatible syntax . Therefore , we are prepared to assert and refine articulations from our concepts to yours for the purpose of maximizing intensional node congruence" . Only then can we expect to also maximize the empirical translatability of biological data linked to diverging phylogenomic hypotheses . Shifting towards the latter attitude will be more challenging than providing the operational logic to enable scalable alignments . Automation of certain workflow components is certainly possible . Ultimately , the logic or technical issues are not the hardest bottlenecks to overcome . Designers of future data environments capable of verbalizing phylogenomic conflict and synthesis need to reflect on how to promote a culture where experts routinely re-/assess the intensionalities of node concepts published by peers . If we wish to track progress and conflict across phylogenomic inferences , we first need to design a value system that better enables and motivates experts to do so . He we discuss various reviewer comments that merit a response but would break up the main flow of the narrative if inserted earlier . We take liberty to assign a header to each comment . Phylogenetic clade definitions and taxonomic concepts are fundamentally mismatched . One reviewer pointed out that clade hypotheses are about branching patterns and relationships of descent , and therefore are mismatched with our notion of node intensionality . We disagree in the following sense . We believe that we are not conflating two fundamentally different kinds of clade conceptualizations , as much as bringing out with the RCC–5 alignments one aspect in the dual , or hybrid nature of clade concepts . The latter are not either this or that–with parallels to the taxa as classes-versus-individuals literature–both can be both , depending on the pragmatic interest [36 , 37 , 62] . For the purpose of synthesis and integration , modeling the intensional aspect of clade concepts is critical . We see this purpose reflected ( e . g . ) in the matching of high-level terminals in [3] . No mechanism for quantitatively expressing uncertainty about tree topology . The same reviewer pointed out that we select single point estimate topologies for each author team , thereby not accounting for the complex likelihood surfaces of the reconstructions and the relative uncertainty of each topology . Applied to what we show here , this criticism is valid . However , it would be feasible perform RCC–5 alignments on a cluster of paired topology alternatives with similar likelihood values . The products can be compared in order to manage uncertainty , through identification of stable versus variable regions across multiple alignments . If most of the variation occurs at higher levels , this would mean that the vast majority of our low-level RCC–5 input articulations could be reused . Phylogenetic conflict is not limited to two trees . Another reviewed pointed out the need to align more than two phylogenies in situations where many recent reconstructions are available to inform a synthesis [5 , 6 , 11] . While the current logic toolkit handles three or more input trees in principle , there certainly are unrealized opportunities to model transitive relationships ( example: for concepts A , B , C in the input trees T1 , T2 , T3: if AT1 = = BT2 and BT2 = = CT3 then AT1 = = CT3 ) . 'Smartly' breaking down alignments of three or more trees while exploiting transitive relationships , as well as visualizing the outcomes accessible ways , are important future improvements for this approach . "Not every clade [concept] is worth labeling and discussing" . We can agree with that assessment . But , having a framework to do so is critical to evaluating the feasibility of a phylogenomic data knowledge service , and should not trail behind discussions regarding its desirability . If we have no formalized means of translating Fig 1 of [3] into a machine-accessible language ( Fig 11 ) , then we cannot fully understand the costs and benefits of building the service . Incentivizing alignment production . One reviewer pointed out that efforts to align multiple trees are costly , and inquired about our suggestions for incentivizing such expert contributions . An initial answer would point to the creation of an e-journal , where multi-phylogeny and -taxonomy alignments can be published either as stand-alone articles or in association with separate publications of new tree reconstructions . The platform of a formal journal best responds to expert needs to receive academic credit [63] . Knowledge systems such as [64] could represent the information input and output . The most valuable product of such an e-journal are the expert-vetted sets of RCC–5 articulations , which represent a new kind of "systematic intelligence" . Scientists and commercial publishers may utilize this intelligence to improve the precision and recall of systematically structured data [54] , where business models would focus on the latter clients for revenue . Needless to say , these are ideas that will take time to concretize and test . | Synthetic platforms for phylogenomic knowledge tend to manage conflict between different evolutionary reconstructions in the following way: "If we do not agree , then it is either our view over yours , or we just collapse all conflicting node concepts into polytomies" . We argue that this is not an equitable way to realize synthesis in this domain . For instance , it would not be an adequate solution for building a unified data environment where authors can endorse and yet also reconcile their diverging perspectives , side by side . Hence , we develop a novel system for verbalizing–i . e . , consistently identifying and aligning–incongruent node concepts that reflects a more forward-looking attitude: "We may not agree with you , but nevertheless we understand your phylogenomic inference well enough to express our disagreements in a logic-compatible syntax . We can therefore maximize the translatability of data linked to our diverging phylogenomic hypotheses" . We show that achieving phylogenomic synthesis fundamentally depends on the application of trained expert judgment to assert parent node congruence in spite of incongruently sampled children . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"taxonomy",
"linguistics",
"social",
"sciences",
"neuroscience",
"animal",
"phylogenetics",
"research",
"design",
"phylogenetics",
"data",
"management",
"cognitive",
"psychology",
"phylogenetic",
"analysis",
"zoology",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"constraint",
"relaxation",
"computer",
"and",
"information",
"sciences",
"syntax",
"sequence",
"alignment",
"bioinformatics",
"evolutionary",
"systematics",
"grammar",
"psychology",
"reasoning",
"database",
"and",
"informatics",
"methods",
"biology",
"and",
"life",
"sciences",
"evolutionary",
"biology",
"cognitive",
"science"
] | 2019 | Verbalizing phylogenomic conflict: Representation of node congruence across competing reconstructions of the neoavian explosion |
The Drosophila heart tube represents a structure that similarly to vertebrates' primary heart tube exhibits a large lumen; the mechanisms promoting heart tube morphology in both Drosophila and vertebrates are poorly understood . We identified Multiplexin ( Mp ) , the Drosophila orthologue of mammalian Collagen-XV/XVIII , and the only structural heart-specific protein described so far in Drosophila , as necessary and sufficient for shaping the heart tube lumen , but not that of the aorta . Mp is expressed specifically at the stage of heart tube closure , in a polarized fashion , uniquely along the cardioblasts luminal membrane , and its absence results in an extremely small heart tube lumen . Importantly , Mp forms a protein complex with Slit , and interacts genetically with both slit and robo in the formation of the heart tube . Overexpression of Mp in cardioblasts promotes a large heart lumen in a Slit-dependent manner . Moreover , Mp alters Slit distribution , and promotes the formation of multiple Slit endocytic vesicles , similarly to the effect of overexpression of Robo in these cells . Our data are consistent with Mp-dependent enhancement of Slit/Robo activity and signaling , presumably by affecting Slit protein stabilization , specifically at the lumen side of the heart tube . This activity results with a Slit-dependent , local reduction of F-actin levels at the heart luminal membrane , necessary for forming the large heart tube lumen . Consequently , lack of Mp results in decreased diastolic capacity , leading to reduced heart contractility , as measured in live fly hearts . In summary , these findings show that the polarized localization of Mp controls the direction , timing , and presumably the extent of Slit/Robo activity and signaling at the luminal membrane of the heart cardioblasts . This regulation is essential for the morphogenetic changes that sculpt the heart tube in Drosophila , and possibly in forming the vertebrates primary heart tube .
During early development , the vertebrate heart exhibits genetic and morphological similarities to the cardiac tube ( dorsal vessel ) of the invertebrate model organism Drosophila melanogaster [1]–[3] . The primary genetic network that determines the heart field has been established in both vertebrates and invertebrates [2] . However , the structural components downstream of this primary transcriptional network inducing the large cardiac tube morphology , which differs significantly from that of the aorta , remain to be elucidated . The Drosophila dorsal vessel is a single tube , formed by the coalescence of two opposing rows of cardioblasts at the dorsal midline [4] . Following their initial encounter , opposing pairs of cardioblasts contact each other by establishing adherens junctions along the dorsal midline . Subsequently , their future luminal membrane curves inward , creating rows of crescent-shaped cardioblasts . Finally , the ventral-most luminal membrane seals the dorsal vessel tube by forming adherens junctions with opposing cardioblasts and the lumen is formed ( Fig . 1 , upper panel ) [5] . The volume of the lumen of the dorsal vessel depends primarily on two parameters , the length and position of dorsal and ventral adherens junctions formed between pairs of opposing cardioblasts , and the extent of the curvature of the luminal membrane . Importantly , the dorsal vessel is divided into two compartments: the non-contractile anterior aorta , which exhibits an extremely narrow lumen , and the contractile heart domain , characterized by a significantly larger lumen [4] . The genes involved in determining the shape and size of the unique lumen of the heart tube have yet to be characterized . Previous genetic analyses have identified multiple components contributing to the morphology of the entire dorsal vessel , including both the heart and the aorta compartments . These include Cadherin-mediated adhesion molecules [5] , Integrin and its extracellular matrix ( ECM ) ligand , Laminin [5]–[8] , Dystroglycan ( Dg ) [9] , elements of the Slit/Roundabout ( Robo ) signaling pathway [8]–[11] , and the Uncoordinated-5 Netrin receptor [12] . However , all these genes are expressed and required for morphogenesis of the entire dorsal vessel , and thus their activity cannot explain heart-specific morphology . Slit/Robo signaling has been proposed to play a key role in promoting lumen formation by inhibiting cadherin-mediated adhesion [9] , [10] . However both proteins are expressed ubiquitously throughout the dorsal vessel , and therefore their activity per se also cannot explain heart-specific morphogenesis . Thus , a structural gene , expressed during the time window of heart tube formation , and localized uniquely at the heart compartment of the dorsal vessel is predicted to mediate heart-specific morphogenesis . We identified the ECM component , Multiplexin ( Mp ) , a Drosophila orthologue of mammalian Collagen XVIII and Collagen XV , as a unique structural protein expressed by the heart but not the aorta cardioblasts of the dorsal vessel . Therefore Mp could be involved in the regulation of heart-specific features . Mp is a secreted multi-domain protein with N-terminal sequences containing Thrombospondin repeats , followed by multiple Collagen repeats , a Collagen trimerization domain , and a C-terminal Endostatin cleavable domain . Mp was described previously in the context of the nervous system and wing development [13] , [14]; however , its function in the heart was never been elucidated . Homozygous mp mutants are viable but exhibit axonal pathfinding defects detected in the embryonic PNS [13] . Here , we describe a specific role for Mp in Drosophila heart tube formation . Our results demonstrate that Mp protein is expressed by the heart , and not the aorta cardioblasts , precisely at the time window of lumen formation . Furthermore , we show that Mp is necessary for heart lumen formation and is sufficient to transform the shape of the aorta tube lumen into that of the heart . Mp executes this function by enhancing Slit/Robo activity presumably by increasing Slit protein stabilization , at the luminal aspects of the heart compartment of the cardiac tube , precisely during the time window of cardiac lumen formation . This leads to the asymmetrical distribution of F-actin along the heart cardioblast membrane essential for the development of the large heart tube lumen . Importantly , the resulting narrow heart tube of mp homozygous mutant flies exhibits a physiological defect , as their heart is less efficient in pumping the hemolymph . The results presented here reveal a novel mechanism essential for shaping the large contractile heart tube . Given the remarkable conservation of signaling pathways involved in heart development between flies and vertebrates , and the expression of both collagen XV [15]–[17] and Slit [18] , [19] in vertebrates heart it is likely that this mechanism also plays a role in morphogenesis of the vertebrate heart tube .
The mRNA of mp is expressed in heart cardioblasts at stage 16 at the same time window as cardiac tube formation [13] . In order to analyze Mp protein distribution in the heart , we generated an antibody against the Endostatin domain ( see Experimental Procedures ) . The specificity of the antibody was demonstrated by its lack of reactivity with homozygous mp mutants ( Supplemental Fig . S1C , C′ ) . Importantly , the antibody staining showed that Mp was specifically localized at the heart lumen , but not at the cardioblast basal domain ( Fig . 1A , A′ , B , B′ ) . The antibody staining labeled also the border between the hindgut endoderm and the surrounding visceral muscles , glial cells in the CNS , and tendon cells ( data not shown ) . Importantly , the anti Mp antibody did not detect staining at the aorta domain ( Fig . 1A , C , C′ ) , although we cannot exclude that the aorta cardioblasts express very low levels of this protein . In addition , Mp was not expressed by the pair of ostia cells ( Fig . 1A , arrowhead ) . Importantly , Mp was only detected in cardioblasts at the stage of tube formation , and not in earlier stages ( Supplemental Fig . S1A , B ) , Thus , Mp exhibits a polarized , heart-specific lumen distribution , during cardiac tube formation . Analysis of cross sections of the heart from homozygous mp mutants revealed a significantly smaller cardiac lumen as compared to controls ( Fig . 1D and E ) . The cross section area and the perimeter of the cardiac lumen of mp mutants were about 55% ( p = 8 . 6E-06 ) , and 27% smaller than the wild type ( p = 3 . 8E-05 ) , respectively ( Fig . 1J , K ) . The mp mutant cardioblasts appeared slightly smaller than wild type cardioblasts ( 13% reduction in cardioblast cell area , p = 0 . 02 ) ( Fig . 1L ) , but could not explain the significantly reduced lumen size . No difference in lumen size between the aorta of wild type versus that of mp mutant was detected ( Fig . 1G and H ) . Thus , the Mp-polarized and specific distribution in the heart lumen , together with the aberrant cardiac morphology in mp mutants led us to conclude that Mp function is specific to the future contractile heart compartment of the dorsal vessel , where it is essential for promoting the unique cardiac tube morphology . To determine whether Mp is sufficient to promote changes in cardiac tube morphology , we overexpressed Mp in the entire dorsal vessel using the muscle-specific driver , mef2-GAL4 . The distribution of the overexpressed Mp remained polarized at the luminal surfaces ( Supplemental Fig . S1D , D′ ) , and it was found to induce a significantly larger cardiac lumen ( Fig . 1F ) , or two ectopic lumens ( data not shown ) . The percentage of embryos showing the large/abnormal lumen was 68% ( n = 25 ) . Overexpression of Mp in the aorta led to the enlargement of the aorta lumen to a size comparable to that of the heart ( Fig . 1I , Supplementary Fig S1E , F ) as detected in 50% of the embryos ( n = 8 ) . These phenotypes suggest that high expression of Mp is sufficient to promote enlargement of the luminal membrane domain in both the heart and the aorta compartments . Taken together , our results suggest that the expression of Mp in the heart luminal domain is sufficient for the induction of the unique shape of the heart cardioblast , and can essentially transform the shapes of the aortal lumen into that typically observed only in the heart compartment . The narrow cardiac lumen observed in mp mutant hearts resembled those that form in slit mutants [9] . We therefore tested for genetic interaction between mp and slit or robo . The lumen area of the heart tube of slit/+;mp/+ double heterozygotes was 48% smaller relative to the wild type heart ( p = 0 . 002 , n = 11 ) ( Fig . 2A , B ) . Moreover , embryos that were double heterozygous for mp and robo exhibited a lumen area that was 79% smaller than wild type heart ( p = 2 . 9×10−9 , n = 9 ) ( Fig . 2C ) . Similarly , the lumen perimeter of robo1/+;mp/+ was 47% smaller than wild type ( n = 9 ) , and that of sli2/+;mp/+ was 32% smaller ( n = 11 ) than wild type perimeter . In addition no significant differences were detected between the overall size of the cardioblasts of the double heterozygous relative to wild type ones . Control embryos heterozygous for each mutation alone showed wild type morphology of the cardiac tube ( Fig . 2 D , E , F ) . Based on these results , we concluded that Mp synergizes with Slit and Robo in its function in lumen formation . To determine whether Mp associates with Slit or Robo in a single protein complex , we next tested whether these proteins co-precipitate from extracts of Drosophila Schneider S2 cells . We transfected S2 cells with Mp , Slit and Robo cDNAs , and immunoprecipitated Slit with anti Slit antibodies from a soluble extract of the transfected cells . Western blot analysis showed that Slit was specifically immunoprecipitated , and as expected Robo was co-immunoprecipitated as well . In addition , Mp specifically co-precipitated with the Slit/Robo complex ( Fig . 2J ) , suggesting that it forms a protein complex with either Slit , Robo , or with both . Complexes immunoprecipitated with anti Robo antibodies also contained Mp in the co-precipitated material ( data not shown ) . Furthermore , immunoprecipitation of Slit from protein extracts of mp mutant embryos , or embryos overexpressing Mp in muscles ( using the mef2-GAL4 driver ) , showed a significant elevation of endogenous Slit protein levels in embryos overexpressing Mp and a comparable reduction in Slit levels in mp mutants , relative to control ( Figure 2K ) . This experiment suggests that Mp might enhance Slit protein stability when both are co-expressed . Interestingly , reaction of the Slit immunoprecipitated material with anti Mp showed a clear band of ∼39 kDa in the lane corresponding to embryos overexpressing Mp , consistent with the association of Slit with the Endostatin domain of Mp . This band was not detected in mp mutants , nor in the control yellow/white embryos , possibly because the levels of Mp in the latter embryos are too low for antibody detection . Full length Mp was not detected in any of the IPs with Slit , however the antibody exhibited high background in this region , so we cannot exclude the presence of full length Mp in the IPs . These results are consistent with an effect of Mp on Slit protein stabilization . Consistent with the co-precipitation experiments , Mp co-localized with Slit at the luminal surfaces of the cardiac tube ( Fig . 2G–I ) . Taken together , these data suggest a functional link between Mp , Slit and Robo in promoting heart lumen formation , and are consistent with the notion that all three proteins are components of a protein complex at the cardioblast luminal membrane . To determine whether Mp activity in the heart requires Slit/Robo signaling , we overexpressed Mp in homozygous slit mutants . slit mutant cardioblasts lack a well defined lumen ( 66% ) , or do exhibit a very small lumen ( 33% ) . Staining for beta catenin/Armadillo in slit mutant heart often shows a straight line between opposing cardioblasts ( Fig . 3 , B , B″ arrowhead ) and Dg is missing from that region ( Fig . 3B′ , arrowhead ) , whereas in the wild type heart , beta catenin/Armadillo is enriched at the dorsal and ventral spots where adherens junctions are formed ( Fig . 3 A , A″ arrowheads ) , and Dg is detected along the entire luminal membrane ( Fig . 3A′ , arrowheads ) [9] . Overexpression of Mp in the wild type embryo led to extension and increased curvature of the luminal membrane ( Fig . 3C–C″ , arrowheads ) , an effect that was similar to overexpression of Robo ( Fig . 3D–D″ ) . However , overexpression of Mp in slit mutant cardioblasts significantly compromised its ability to induce an extended , curved luminal membrane . Either only a very small lumen was detected ( in 86% of the mutant embryos ( n = 14 ) ) ( Fig . 3E–E″ , arrowhead ) , or no lumen was formed between opposing cardioblasts . This suggests that Mp-dependent enlargement of the cardiac tube lumen requires Slit/Robo signaling . Interestingly , although Mp-dependent enlargement of the heart lumen was compromised , Mp was able to partially inhibit accumulation of Armadillo/beta-catenin at the luminal membrane ( detected in slit mutants ) , and a concomitant membrane distribution of Dystroglycan reappeared ( Fig . 3E′ , E″ ) . These results suggest an additional effect of Mp that is Slit independent . We detected multiple cytoplasmic Armadillo vesicles in the cardioblast cytoplasm following overexpression of Mp , or Robo ( Fig . 3 C″ , D″ , white filled arrows ) , but not in wild type , slit , or slit overexpressing Mp cardioblasts ( Fig . 3 A″ , B″ and E″ ) . Taken together these results suggest that Mp activity in promoting a large heart lumen requires Slit/Robo signaling . In wild type heart of a stage 16 embryo , Slit is detected in small cytoplasmic patches , often localized close to the luminal membrane ( Fig . 4A , A′ ) . These Slit patches were not described previously in wild type embryos , and could be clearly identified by our new method of visualizing the cardioblasts in cross sections [35] . We found that the number and size of these Slit patches correlated with the extent of Slit/Robo signaling; In robo−/− cardioblasts the average fluorescent intensity of Slit patches per single cardioblast ( calculated as described in M&M ) , was 36% of that of wild type ( p = 0 . 0015 , n = 7 ) ( Figure 4B , B′ ) , whereas in cardioblasts overexpressing Robo , Slit average fluorescent intensity was 2 . 24 fold higher that that of wild type ( p = 0 . 0005 , n = 19 ) ( Figure 4C , C′ ) . Thus , the average fluoresence of cytoplasmic Slit patches represents a measure for the extent of Slit/Robo activity , and can be used to quantify the degree of Slit/Robo signaling . To verify that the Slit cytoplasmic patches represent endocytic vesicles formed as a result of Slit-Robo interaction at the luminal membrane , we partially arrested endocytosis by cardioblast-specific expression of a dominant-negative version of Rab-5-YFP . This led to detection of slightly larger Slit vesicles , at the luminal membrane , and importantly these vesicles overlapped Rab5-YFP labeling ( Fig . 4F–F′″ ) . The lack of these Slit endocytic vesicles in robo mutants , and the elevation in their size and number following Robo overexpression further supported the idea that the Slit cytoplasmic “patches” represent endocytic Slit vesicles formed following the binding of Slit to Robo at the luminal membrane . Unfortunately , we could not test for co-localization of Robo and Slit in a given vesicle because both anti Slit and anti Robo antibodies were made in mouse . Significantly , in mp mutant cardioblasts , the average fluorescence of cytoplasmic Slit vesicles was reduced to 61% than that of wild type ( p = 0 . 009 , n = 30 ) ( Figure 4D , D′ ) , and Slit distribution was mainly detected at the luminal membrane , overlapping that of Dystroglycan ( Dg ) ( Fig . 4D , D′ ) , supporting the idea that Slit/Robo activity was attenuated . Reciprocally , overexpression of Mp resulted in a 1 . 8 fold increase of Slit average fluorescence intensity relative to that of wild type ( p = 0 . 0003 , n = 27 ) ( Figure 4E , E′ ) . Consistent with the lack of Mp in the aorta , we found fewer cytoplasmic Slit vesicles in this domain ( Figure 4G , G′ ) , and Slit was mainly detected at the luminal membrane , overlapping the distribution of Dg similarly to its distribution in mp mutant heart cardioblasts ( Fig . 4G , G′ ) . However , Mp overexpression in the aorta often led to elevation in the number and size of cytoplasmic Slit vesicles correlating with the enlargement of the aorta lumen ( Figure 4 , H , H′ ) . Based on the phenotypic similarity between Mp and Robo activities , both in the induction of a large heart lumen , as well as in promoting the formation of Slit cytoplasmic vesicles it was concluded that Mp represents a positive element in the Slit/Robo signaling pathway . Taken together , these results strongly support our hypothesis that the presence of Mp at the luminal surfaces of the heart tube enhances Slit/Robo activity in a polarized fashion at the luminal domain , and eventually promotes the formation of a large cardiac tube lumen . Our results so far demonstrate a tight functional and molecular link between Mp and Slit/Robo signaling in cardiac morphogenesis . To elucidate whether Mp is capable of modulating Slit/Robo signaling in other tissues , and to differentiate between an effect of Mp on the Slit ligand or on its receptor , Robo , we examined the consequences of overexpression of Mp in midline glia cells , which endogenously express Slit [20] , but not Robo . Staining of the longitudinal CNS axons with anti Fasciclin II ( FasII ) in wild type embryos shows three longitudinal axonal tracks on each side of the midline ( Fig . 4I ) . Previous analysis demonstrated that the distance of these tracks from the midline , as well as their position relative to each other , depends on the level of Robo receptors expressed by the axons of each track [21] , [22] . Strikingly , overexpression of Mp in midline cells ( using the sim-GAL4 driver ) reduced the distance of the longitudinal axons from the midline , and occasionally led to their defasciculation , forming four tracks ( Fig . 4I′ ) . Both phenotypes are consistent with the attenuation of Slit's repulsive activity from the midline , implying a negative effect of Mp on Slit , rather than on Robo . Cross sections of the ventral cord of these embryos , and their labeling with anti Slit antibodies indicated an unusual accumulation of endogenous Slit surrounding the midline glia cells ( Fig . 4J–K′ , arrowheads ) , suggesting that ectopic Mp promoted Slit accumulation surrounding the midline glia cells . The midline glia cells are marked with anti Held Out Wing ( HOW ) shown previously to label these cells [23] . We therefore propose that Mp association with Slit at the surfaces of the midline glia cells inhibits Slit diffusion from the midline glia cells , thus phenocopying a slit hypomorphic phenotype . Although the end result of Mp activity in the CNS is opposite to its effect in the heart , in both cases Mp was capable of modulating Slit distribution , presumably by the association of both proteins in a single protein complex , further supporting our hypothesis that by the formation of an Mp-Slit protein complex , Mp elevates local accumulation of Slit , which in the heart , leads to increased binding to Robo . In axons , Slit/Robo signaling counteracts actin polymerization , leading to membrane retraction [24]–[26] . Therefore , we investigated the relative levels of F-actin in the luminal versus basolateral surfaces of heart cardioblasts . At stage 15 , the entire cardioblasts surface showed positive phalloidin staining ( Fig . 5A , A′ ) . At stage 16 , when the cardiac lumen has been established , we detected a significant reduction of F-actin at the luminal surfaces , while the basolateral levels of F-actin remained high ( Fig . 5B , B′ ) . Significantly , in both slit , and mp mutant embryos , there was no difference between the luminal and basal F-actin levels as observed in wild type ( Fig . 5C , C′ , D , D′ ) . This difference could not be attributed to the close proximity of opposing luminal membranes because it was observed even when a partial small lumen had been formed in mp mutant embryos ( Fig . 5D′ ) . Furthermore , overexpression of Mp in cardioblasts led to extension of the luminal membrane as well as a significant reduction of F-actin at the luminal domain ( Fig . 5E , E′ arrow ) , similar to the effect obtained following overexpression of Robo ( Fig . 5F , F′ arrow ) . In both cases , reduced F-actin correlated with the formation of an extended luminal membrane . Importantly , overexpression of Mp in slit mutant cardioblasts did not lead to reduction of luminal F-actin relative to its basolateral levels ( as in wild type hearts ) , suggesting that this reduction is dependent on Slit ( Fig . 5G , G′ ) . These results demonstrate that the Slit-dependent Mp activity in promoting enlargement of the heart lumen compartment correlates with a significant inhibition of F-actin polymerization along the luminal membrane . Finally , we determined whether the morphological defects observed in mp mutant embryos are of physiological significance for the functional contracting heart of the adult fly . The heart in flies does not undergo extensive histolysis , and the embryonic cardioblasts maintain their identity throughout development . Mp distribution was maintained in the adult fly heart , where it could be detected both along the luminal membrane , as in the embryo ( Fig . 6A , A′ , white empty arrowheads ) , as well as at the contact area between the ventral muscles and the cardiac tube ( Fig . 6A , A′ , white filled arrowhead ) . In both cases it overlapped the Laminin distribution . The physiological relevance of the aberrant heart morphology detected in the embryonic heart of mp mutants was analyzed by measurements of live contracting hearts of mp mutants as well as of wild type adult hearts [27] ( see also representative videos in Supplementary Material , movie S1 ) . We were specifically interested in the effects of mp on heart contractility , given the reduction in lumen size observed in mp mutants . This can be quantified as the percent fractional shortening , which is essentially the extent of contraction during systole compared to diastole ( [diastolic diameter-systolic diameter]/diastolic diameter ) . We observed ( Fig . 6B ) a significant reduction ( 25% versus 37% P<0 . 005 ) in the percent fractional shortening in mp mutants , indicating reduced contractility of the heart tube . This is also evident in M-mode records from these hearts , which show the movement of the heart walls ( y-axis ) over time ( x-axis ) . These movements are much shallower in mp mutant hearts , resulting from a decreased ability of the heart walls to retain full diastolic values ( Fig . 6C ) . A video of representative heart contraction of mp mutant fly is shown in Supplementary information ( movie S2 ) . Therefore , these results indicate that Mp is essential for proper heart contraction , and in its absence , the pumping efficiency of the hemolymph in the adult fly is compromised .
Because the aortal domain develops a lumen ( although small ) in the absence of Mp , we favor the idea that while Slit/Robo signaling is critical for lumen formation per se , Mp activity modulates this signaling in a polarized fashion , eventually leading to higher level of activation of Slit/Robo signaling at the heart luminal membrane . The precise time window of Mp expression , which correlates with the timing of cardiac lumen formation and its polarized luminal distribution are both critical for cardiac-specific tube morphogenesis , as both Slit and Robo distributions are wider and extend along the entire dorsal vessel . The significant reduction in Slit endocytic vesicles in mp mutants and their enhanced appearance following Mp overexpression implicate Mp as a critical factor in promoting the constitutive local activation of Slit/Robo signaling . Unlike the situation in the nervous system , Slit in the cardioblasts acts in an autocrine , cell autonomous manner [9] , [10] , and its activity appears to be continuous ( as judged by its multiple endocytic vesicles ) . We therefore suggest that Mp , expressed in a tightly controlled manner in the heart , is required for regulating Slit activity . Excessive signaling leads to an aberrant , often opened tube as observed in Mp overexpression , and was also reported following Robo overexpression [10] . On the other hand , reduced Slit signaling leads to a very narrow tube as seen in distinct robo , mp or slit mutants . The role of Mp as a heart-specific local enhancer of the Slit/Robo signaling pathway is supported by the following findings: ( a ) It co-localizes with Slit on the heart luminal surfaces . ( b ) It forms a protein complex with Slit in S2 cells and in embryos . ( c ) It appears to enhance Slit protein stability . ( d ) In its absence , Slit endocytic vesicles are significantly reduced ( as in the robo mutant ) . ( e ) Mp overexpression leads to supernumerary Slit endocytic vesicles ( similar to the Robo overexpression phenotype ) . ( f ) Its ability to induce an extended cardiac tube lumen is suppressed in slit mutants . ( g ) It facilitates Slit accumulation in the midline following its overexpression . The effect of Mp on Slit might be explained on a molecular level by local enhancement of Slit protein stability , demonstrated by elevation of endogenous Slit levels , and/or accumulation following Mp overexpression . Slit binding to chondroitin-sulfate chains of Syndecan has been shown to increase its affinity for Robo [28] . Mp proteins contain chondroitin sulfate chains [14] , and thus could bind directly to Slit and Robo via these chains . This can promote local accumulation of Slit at the luminal membrane and enhance Slit/Robo interactions . Alternatively , Mp might facilitate the interaction between Slit and Syndecan , which was previously shown to contribute to heart morphology [29] . The latter possibility is less likely because we did not find genetic interaction between syndecan and mp in double heterozygous embryos and the hearts of the latter was normal ( N . Harpaz , data not shown ) . Our results suggest that whereas Slit is epistatic to Mp , overexpression of Robo in the aorta is capable of promoting expansion of the aorta , even in the absence of Mp ( data not shown ) , implicating that Mp is not absolutely required for Slit/Robo activation . Previous studies suggested that the Slit/Robo signaling pathway counteracts the formation of adherens junctions at the luminal surfaces [9] , [10] . However , the ubiquitous distribution of Slit along both the aorta and heart domain , and its earlier function in promoting the polarity of the cardioblasts prior to the formation of the tube [11] , cannot explain the direct involvement of Slit in the specific formation of the heart lumen . Mp , on the other hand , is secreted to the luminal ECM during tube closure and lumen formation , and therefore most likely plays a specific role during this stage . Our results are consistent with a major function for Slit/Robo signaling in reducing F-actin levels at the luminal surfaces , thus leading to asymmetrical distribution of F-actin along the cardioblast membrane . Asymmetric distribution of F-actin as a result of cell confinement or differential adhesion has been recently shown to promote spontaneous lumen formation in MDCK cells [30] , supporting our observations . In addition , reduced F-actin at the luminal membrane might have a dual function for cardiac tube lumen formation: It could restrict the domain of dorsal and ventral adherens junctions , allocating more luminal membrane at the expense of junctional membrane , and it may facilitate the inward curvature of the luminal membrane . This latter effect is probably tightly controlled , as extended curvature leads to the formation of lumen inside a single cardioblast as observed in cells overexpressing Mp . The inhibitory effect of Mp on beta-catenin was deduced from the accumulation of Armadilo vesicles following Mp overexpression , as was similarly observed following Robo overexpression . In vitro studies previously linked Slit/Robo signaling to reduced N-cadherin adhesion resulting in detachment of beta-catenin from N-cadherin [31] . A direct , Slit-independent effect of Mp on beta-catenin is possible , as a recent study showed that Mp promotes a decrease of the extracellular Wg/Wnt protein in the Drosophila proventriculum zone [14] and the C-terminal cleaved polypeptide Endostatin has consistently been shown to inhibit Wnt signaling , by promoting beta catenin degradation in vertebrates [32] . Both mammalian Mp orthologue genes , Collagen XV and Collagen XVIII , have been linked to various diseases . Collagen XV had been recently associated with predisposition to cardiomyopathy [16] , [17] and similarly the Slit/Robo signaling pathway has been linked to cardiac morphogenetic defects [19] , it is possible that these two proteins functionally interact to promote cardiac morphogenesis . A Collagen XVIII deficiency leads to Knobloch syndrome , which is characterized by various pathologies , including viteoretinal degeneration with retinal detachment as well as a neural tube closure defect [33] , [34] . Whereas no heart defects were described in the col18a mutant mice , or in humans with Knobloch syndrome , a possible functional connection between ColXVIII and the Slit/Robo signaling pathway might exist . Finally , similarly to Collagen IV shown to modulate TGF-β signaling , [35] , [36] , Mp appears to modulate Slit signaling by controlling its levels . In summary , our analysis revealed a novel mechanism of cardiac tube lumen morphogenesis associated with the polarized tissue-specific activity of mp , the Drosophila orthologue of the col18a and col15 genes . Our study explains how Mp interaction with the Slit/Robo signaling pathway promotes the inward curvature of the luminal membrane specifically in the heart domain .
Wild type flies were yw or hand-GMA-GFP ( obtained from F . Schnorrer , MPI , Martinsried , Germany ) , mp mutants were dmpΔN4-11 , and UAS-Mp-3hnc1 ( obtained from B . Moussian , MPI , Tubingen , Germany ) and were described in [13] , sli2/CyO , Yfp and robo1/CyO , Yfp obtained from B . J . Dickson , Research Institute of Molecular Pathology , Austria , and UAS-Robo was obtained from SG . Kramer , UMDNJ-Robert Wood Johnson Medical School described in [10] . GAL4 lines were mef2-GAL4 and sim-Gal4 ( Bloomington stock center , Indiana , USA ) . For the induction of Rab5-dominant negative , a UAS-Rab5-DN-YFP line was obtained from the Bloomington stock center , Indiana , USA . For the induction of Mp over expression , two copies of the mef2-GAL4 and two copies of the UAS-Mp were used ( at 25°C ) , with the exception of Fig . 1K , where one copy of each construct was used but the embryos were raised at 29°C . For Robo over expression , single copies of mef2-GAL4 and UAS-Robo were used . Selection of the sli2 and robo1 homozygous embryos was done on the basis of a YFP-containing balancer . The selection of sli2 mutant embryos overexpressing Mp , was based on an aberrant CNS phenotype . Primary antibodies used included: rabbit anti Mef2 ( 1∶200 , Nguyen H . T , University of Erlangen-Nuremberg , Germany ) , rat anti Mp ( 1∶100 for whole embryos , 1∶35 in sections ) was produced in our lab by rat immunization of a purified GST fused to the Endostatin domain of Mp , mouse anti Fasciclin II ( 1∶10 , this monoclonal antibody was developed by Goodman , C . , was obtained from the Developmental Studies Hybridoma Bank , developed under the auspices of the NICHD and maintained by The University of Iowa , Dept of Biology , Iowa City , IA 52242 ) ; rabbit anti Dg ( 1∶140 , a gift from W . M . Deng , Florida State University , FL , USA ) [37] , rat and rabbit anti HOW ( 1∶35 , were produced in our lab ) , guinea pig anti Laminin ( 1∶70 , produced in our lab ) , mouse anti Slit ( 1∶10 , this monoclonal antibody was developed by Artavanis-Tsakonas , S . , was obtained from the Developmental Studies Hybridoma Bank , developed under the auspices of the NICHD and maintained by The University of Iowa , Dept of Biology , Iowa City , IA 52242DSHB ) , mouse anti Armadillo ( 1∶70 , this monoclonal antibody was developed by Wiescjaus , E . was obtained from the Developmental Studies Hybridoma Bank , developed under the auspices of the NICHD and maintained by The University of Iowa , Dept of Biology , Iowa City , IA 52242DSHB ) , chick anti GFP ( 1∶70 , Aves Labs , Inc ) . Alexa Fluor 488 phalloidin was used ( 1∶140 , Invitrogen , A12379 ) for F-actin labeling . Secondary antibodies conjugated with Cy2 , Cy3 , Cy5 or 405 raised against guinea pig , rat , rabbit , mouse and chick were purchased from Jackson Laboratories , Inc . Staged embryos were collected , dechorionated and fixed as described previously [38] . For phalloidin staining EtOH instead of MeOH was used during fixation , otherwise embryos were fixed as described in Gilsohn and Volk [39] . Confocal images were taken with a Zeiss LSM710 confocal system , and processed with Adobe Photoshop . The Fiji/ImageJ program was used to quantify the fluorescence intensity of cytoplasmic Slit in cardioblasts . A threshold for vesicles fluorescence was defined ( according to the signal/background ratio ) on average intensity projection lsm images , and the amount of fluorescence within each of the cardioblasts was calculated as the percentage of the vesicles area relative to the total area of the cardioblast ( visualized by Dg staining ) . The fluorescence intensity was normalized to that of wild type cardioblast . The statistical significance of the measurments was calculated by student T test . The analysis was performed as described previously [27] , [40] . In brief , flies ( 1–2 week old ) were dissected and their beating hearts were filmed using high speed digital cameras ( EM-CCD digital camera , Hamamatsu Corp ) and light microscopy . Heart measurements and calculation of physiological parameters were performed with the HC image data capture software ( Hamamatsu Corp ) . The percentage of fractional shortening was calculated as the difference between the diastolic and the systolic ranges , divided by the diastolic range . The result was multiplied by 100 . Lumen and cardioblasts dimensions were measured using the Zen 2008 Confocal program . The measurements were taken from lsm projection images , using the open/closed Bezier tool . Constructs used for cells transfection included: pUAST-robo-HA ( BJ Dickson , Research Institute of Molecular Pathology , Austria ) , pUAST-slit ( GJ . Bashaw , University of Pennsylvania School of Medicine , Philadelphia , USA ) and pUAST-mp ( B . Moussian , Max-Planck Institute , Germany ) . SR+ cells were transfected all 3 cDNA constructs ( Robo , Slit , Mp ) using the Escort IV transfection reagent ( Sigma-Aldrich , # L3287 ) , and grown for 36 hours . The cells were pelleted and lysed in NP40 0 . 5% containing PBS . The soluble lysate was incubated with protein A/G beads ( SC-2003 , Santa Cruz ) pre-coupled with mouse anti Slit antibody or with normal mouse serum as a control , over night at 4°C , washed 3 times with the extraction buffer , and boiled in sample buffer , and separated by SDS polyacrylamide gel electrophoresis . For embryos protein extract analysis , stage 16 embryos were collected and lysed with RIPA buffer . For Western blot analysis , proteins were transferred onto a nitrocellulose membrane , blocked with 5% milk in PTW , reacted with the primary antibody ( 2 hrs , at RT ) , washed in PBSX3 and then reacted with the secondary antibody conjugated to HRP . ECL reaction was performed after multiple washings of the nitrocellulose membrane . | The formation of the characteristic large heart lumen is common to all heart-containing organisms and is essential for efficient heart function; however , the structural components promoting this process are yet to be elucidated . The Drosophila heart represents a specific compartment within an elongated contractile tube , the dorsal vessel , essential for pumping the hemolymph throughout the fly body . Here , we describe a novel extracellular matrix component , Multiplexin ( Mp ) , homologous to vertebrates Collagen XV/XVIII , which is necessary and sufficient for promoting the large heart lumen . Based on molecular and genetic analysis , our findings link Mp activity to a signaling pathway ( Slit/Robo ) demonstrated previously to repress actin polymerization at the leading edge of migrating neurons . Consistently we show that Mp deposited at the luminal membrane enhances Slit/Robo activity and presumably signaling , leading to reduced actin levels , necessary for curving of the luminal membrane , and for the formation of the large heart lumen . Consequently , mp mutant flies exhibit narrow heart and reduced heart contractility . These results demonstrate a novel mechanism by which local deposition of an ECM component promotes a polarized signaling at the luminal aspects of a pair of cardioblasts , shaping the large heart tube compartment . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"developmental",
"biology",
"organism",
"development",
"morphogenesis",
"organogenesis",
"biology",
"anatomy",
"and",
"physiology",
"cardiovascular",
"system"
] | 2013 | Multiplexin Promotes Heart but Not Aorta Morphogenesis by Polarized Enhancement of Slit/Robo Activity at the Heart Lumen |
The BAR domain protein superfamily is involved in membrane invagination and endocytosis , but its role in organizing membrane proteins has not been explored . In particular , the membrane scaffolding protein BIN1 functions to initiate T-tubule genesis in skeletal muscle cells . Constitutive knockdown of BIN1 in mice is perinatal lethal , which is associated with an induced dilated hypertrophic cardiomyopathy . However , the functional role of BIN1 in cardiomyocytes is not known . An important function of cardiac T-tubules is to allow L-type calcium channels ( Cav1 . 2 ) to be in close proximity to sarcoplasmic reticulum-based ryanodine receptors to initiate the intracellular calcium transient . Efficient excitation-contraction ( EC ) coupling and normal cardiac contractility depend upon Cav1 . 2 localization to T-tubules . We hypothesized that BIN1 not only exists at cardiac T-tubules , but it also localizes Cav1 . 2 to these membrane structures . We report that BIN1 localizes to cardiac T-tubules and clusters there with Cav1 . 2 . Studies involve freshly acquired human and mouse adult cardiomyocytes using complementary immunocytochemistry , electron microscopy with dual immunogold labeling , and co-immunoprecipitation . Furthermore , we use surface biotinylation and live cell confocal and total internal fluorescence microscopy imaging in cardiomyocytes and cell lines to explore delivery of Cav1 . 2 to BIN1 structures . We find visually and quantitatively that dynamic microtubules are tethered to membrane scaffolded by BIN1 , allowing targeted delivery of Cav1 . 2 from the microtubules to the associated membrane . Since Cav1 . 2 delivery to BIN1 occurs in reductionist non-myocyte cell lines , we find that other myocyte-specific structures are not essential and there is an intrinsic relationship between microtubule-based Cav1 . 2 delivery and its BIN1 scaffold . In differentiated mouse cardiomyocytes , knockdown of BIN1 reduces surface Cav1 . 2 and delays development of the calcium transient , indicating that Cav1 . 2 targeting to BIN1 is functionally important to cardiac calcium signaling . We have identified that membrane-associated BIN1 not only induces membrane curvature but can direct specific antegrade delivery of microtubule-transported membrane proteins . Furthermore , this paradigm provides a microtubule and BIN1-dependent mechanism of Cav1 . 2 delivery to T-tubules . This novel Cav1 . 2 trafficking pathway should serve as an important regulatory aspect of EC coupling , affecting cardiac contractility in mammalian hearts .
The BAR domain superfamily is composed of proteins involved in endocytosis , organelle biogenesis , cell division , and cell migration ( review in [1] ) . As a member of the BAR domain superfamily , the tubulogenesis membrane scaffolding protein BIN1 ( Amphiphysin 2 ) is known to induce membrane invagination [2] , [3] and initiate tubulogenesis in skeletal muscle cells [4] . BIN1 deforms the membrane bilayer through interaction between its N-terminal positively charged BAR domain and acidic phospholipids within the cell membrane [4] , [5] . Knowledge of the role of BIN1 in muscle cells includes evidence of BIN1 distribution on T-tubules of skeletal myocytes [4] and that constitutive knockdown of BIN1 in mice is perinatal lethal , with pathology revealing a hypertrophic dilated cardiomyopathy [6] , [7] . However , despite these findings , little is known of the functional role of BIN1 in cardiomyocytes . Since BIN1 knockdown results in cardiomyopathy [6] , it is possible that BIN1 may play a role in regulating the cardiac calcium transient . During each heartbeat , calcium release from intracellular stores is achieved when trans-sarcolemmal calcium activates the ryanodine release channels on the sarcoplasmic reticulum ( SR ) [8] . The initial calcium influx occurs primarily through the L-type calcium channels with Cav1 . 2 as the pore-forming subunit . Trans-sarcolemmal calcium entry and activation of ryanodine receptors is a local phenomenon and , in cardiomyocytes , sarcolemmal Cav1 . 2 channels occur within 15 nm of their respective ryanodine receptors on the SR [9] . A major function of T-tubule invaginations of the sarcolemma , which are enriched with Cav1 . 2 channels [10] , [11] , is to bring the channels into close proximity of the ryanodine receptors , amplifying sarcolemmal calcium entry to a large calcium release from the SR . This process , which is known as calcium-induced calcium release ( CICR ) [12] , is essential to each heartbeat and links electrical excitation of the myocyte and local calcium entry to its mechanical contraction . The mechanism for Cav1 . 2 localization to T-tubules remains unknown . It is possible that locally enriched BIN1 may assist in the delivery of Cav1 . 2 channels in a manner similar to the role of adherens junctions in aided delivery of Connexin43 ( Cx43 ) hemichannels to intercalated discs [13] , a highly efficient trafficking pathway for polarized protein distribution . Therefore , depletion of BIN1 at T-tubule membrane after knockdown could result in mislocalization of Cav1 . 2 , causing inefficient excitation-contraction ( EC ) coupling and lethal cardiomyopathy . Supporting evidence for this Cav1 . 2 localization hypothesis is that another BAR domain containing protein , endophilin , has been shown to complex with Cav1 . 2 at the plasma membrane [14] . Furthermore , BIN1 has been shown to interact not only with cortical actin [15] , [16] but also with a microtubule plus end tracking protein [17] . These data indicate that BIN1 might be closely associated with growing microtubules , a key component of the trafficking machinery for targeted delivery . In this study , we provide data supporting a role for BIN1 in tethering microtubules for direct delivery of L-type calcium channels to cardiac T-tubules . We observed that in both human and mouse cardiomyocytes , BIN1 and Cav1 . 2 colocalize at cardiac T-tubules ( by fluorescence and electron microscopy immunogold labeling ) and co-immunoprecipitate . Regarding delivery to T-tubules , we found that BIN1 tethers dynamic microtubules and forward trafficking of Cav1 . 2 channels is microtubule dependent . In reductionist atrial myocyte and non-myocyte cell systems , BIN1 is sufficient to form membrane invaginations and distribute Cav1 . 2 to these BIN1-containing membrane regions ( visualized by total internal reflection microscopy , TIRFm ) . The delivery results in non-myocyte cells suggest that Cav1 . 2 delivery to BIN1 is independent of other myocyte-specific organelles and proteins . To rule out that the sarcolemmal invaginations themselves and not BIN1 are sufficient for Cav1 . 2 delivery , we created C-terminal truncated BIN1 , which fails to attract Cav1 . 2 yet still inducing membrane invagination . In isolated primary mouse ventricular cardiomyocytes , disruption of this delivery mechanism by BIN1 knockdown results in less surface expression of Cav1 . 2 and abnormal calcium transient development . Our findings indicate that the membrane curvature protein BIN1 can form membrane invaginations and is localized to cardiac T-tubules , providing an anchor for microtubules that allows targeted delivery of Cav1 . 2 channels and regulation of the cardiac calcium transient . A role of BIN1 in facilitating microtubule-based antegrade delivery of membrane protein traffic adds an important facet to the multifunctional BAR domain family . Furthermore , our findings suggest that microtubule-based delivery of Cav1 . 2 to BIN1 is significant to cardiac calcium regulation .
To understand the cellular distribution of BIN1 in mammalian cardiomyocytes , we dissociated non-failing human cardiomyocytes from freshly explanted human hearts with normal left ventricular function , as well as normal adult mouse cardiomyocytes . After fluorescence immunostaining , the cardiomyocytes were imaged at Z-depth increments of 0 . 1 µm with a spinning disc confocal microscope and viewed in two-dimensional frame views along the longitudinal axis . BIN1 has a nuclear localization , as previously reported , in embryonic hearts ( Figure S1 ) [6] , but elsewhere in the cardiomyocyte , a typical T-tubule distribution pattern of BIN1 emerges that is similar to Cav1 . 2 distribution ( Figure 1A , first row ) . Representative fluorescence intensity profiles along the longitudinal axis of cardiomyocytes are in the second row of Figure 1A . Note that there is a fluorescence signal peak approximately every 2 µm , which corresponds to the T-tubule distribution of the protein . Power spectrum analysis [18] confirms that the fundamental periodicity of Cav1 . 2 is 2 µm ( Figure 1A , third row ) , which is consistent with previously reported cardiac T-tubule intervals [18] , [19] . BIN1 shows the same spatial periodicity as Cav1 . 2 in human and mouse ( Figure 1A ) cardiomyocytes . Cx43 , which localizes at intercalated discs at the longitudinal ends of cardiomyocytes , does not have the same spatial periodicity and serves as a negative control ( Figure S2 ) . The data of Figure 1A indicate that BIN1 is localized along T-tubules in cardiomyocytes . Next , we quantified colocalization between Cav1 . 2 and BIN1 . In Figure 1B , immunolabeling of BIN1 ( green ) and Cav1 . 2 ( red ) is shown in subsections of both human and mouse cardiomyocytes . Full cardiomyocyte views of colocalization between BIN1 and Cav1 . 2 are shown in Figure S3 . The data indicate that BIN1 significantly colocalizes with Cav1 . 2 , primarily at T-tubules . For negative control studies , Cav1 . 2 does not have significant colocalization with Cx43 ( Figure S4 ) . To confirm spatial coincidence , we used transmission electron microscopy with dual immunogold labeling to identify Cav1 . 2 and BIN1 on T-tubule ultrastructures in adult mouse cardiomyocytes . Results in Figure 1C ( left panel ) indicate that BIN1 ( small 10 nm dots ) and Cav1 . 2 ( large 15 nm dots ) are enriched and occur within 10–50 nm of each other at T-tubular membrane structures . The data in Figure 1C indicate close approximation of Cav1 . 2 and BIN1 in isolated cardiomyocytes but do not reveal how the proteins achieve such localization . There is significant support for membrane ion channel delivery occurring via microtubules [13] , [20] , [21] . To address whether BIN1 serves as a microtubule anchoring site to allow Cav1 . 2 delivery , we first studied microtubule behavior in the vicinity of BIN1 in HeLa cells , which are permissive to high-resolution imaging . HeLa cells were transfected with α-tubulin-GFP and BIN1-mCherry . Introduction of exogenous BIN1 forms membrane invaginations as previously reported in other non-myocyte cell types [4] . Twenty-four hours post-transfection , microtubule dynamics were recorded with spinning disc confocal microscopy for 2 min with a frame rate of 1 s . As seen in the enlarged panel of the overlay between BIN1 ( red ) and microtubules ( black lines ) in Figure 2A , microtubules tether at BIN1 structures . Three representative microtubule travel paths involving a microtubule that remains at BIN1 ( MT1 ) , a microtubule that departs BIN1 ( MT2 ) , and a microtubule that approaches BIN1 ( MT3 ) are also indicated in green . For each of these three microtubules , the distance between the microtubule tip and the center of the closest BIN1 structure is plotted over time in Figure 2B . In each graph , the distance within 0 . 2 µm of the respective BIN1 structure is highlighted in red dotted lines . MT1 has paused at BIN1 structure for the whole 2 min imaging window and has relatively little movement . However , MT2 pauses and hovers at BIN1 and , upon leaving , picks up velocity , while MT3 approaches BIN1 with high velocity before it slows down as it comes into contact with BIN1 . The dynamic movements of MT1 , MT2 , and MT3 are shown in Video S1 . In addition , the overall microtubule dynamics tabulated from 15 microtubules of four individual cells are presented in Table 1 . These data indicate that overall tip velocity is 5× faster ( 0 . 15 µm/s versus 0 . 03 µm/s ) when the microtubules are not in the proximity of BIN1 . This increased overall velocity consists of not only faster growth and shortening velocities but also less frequent pauses . The data from Figure 2 and Table 1 suggest that microtubules are tethered by BIN1 structures . To evaluate if microtubules are involved in antegrade trafficking of Cav1 . 2 channels , we exposed live primary adult ventricular cardiomyocytes to the microtubule disruptor nocodazole in the presence of dynasore , a specific dynamin GTPase inhibitor that blocks endocytosis [22] . Expression of surface membrane-bound Cav1 . 2 was assayed by cell surface biotinylation ( Figure 3A ) . Dynasore treatment alone increases surface expression of Cav1 . 2 , indicating inhibition of Cav1 . 2 endocytosis . In the presence of both dynasore and nocodazole , Cav1 . 2 surface expression progressively decreases , further suggesting that microtubule disruption reduces forward trafficking of Cav1 . 2 to the plasma membrane . To confirm that delivery of Cav1 . 2 to T-tubules is microtubule dependent , the cellular distribution of Cav1 . 2 in cardiomyocytes subjected to nocodazole was studied by immunoconfocal microscopy . As seen in Figure S5 , nocodazole decreases Cav1 . 2 surface expression not only at T-tubules ( Figure S5 , bottom right ) but also at global sarcolemma containing non-T-tubule membrane ( Figure S5 , bottom left ) . The total cellular protein expression level of Cav1 . 2 is not changed by nocodazole ( Western blot in Figure S5 , top panel ) . Microtubule-dependent trafficking of Cav1 . 2 is further supported by the localization of Cav1 . 2 vesicles along the microtubule network in the vicinity of a T-tubule in adult mouse cardiomyocytes ( Figure 3B , top panel ) . To better visualize microtubules and Cav1 . 2 , we used the cardiomyocyte-derived HL-1 cell line that has a morphology amenable to high-resolution imaging [23] and find that Cav1 . 2 distributes along the microtubule network ( Figure 3B , bottom panel ) . Comparable biotinylation results confirm microtubule-dependent surface expression of Cav1 . 2 in HL-1 cells ( Figure S6 ) . From the data in Figures 1–3 , it appears that BIN1 is enriched along cardiac T-tubules and closely associated with Cav1 . 2 . Furthermore , BIN1 tethers to plasma membrane dynamic microtubules , which deliver Cav1 . 2 to the plasma membrane . Therefore , it is possible that BIN1 is a T-tubule anchor site for targeted delivery of Cav1 . 2 through the interaction between BIN1 and growing microtubules . To test the exclusivity of the relationship between Cav1 . 2 and BIN1 , we explored whether Cav1 . 2 could be targeted to exogenous BIN1-induced membrane invaginations in cell lines lacking a developed T-tubule system . HL-1 cells are myocytes that express endogenous Cav1 . 2 but do not have a developed T-tubule system . Introduction of exogenous BIN1 generates membrane invaginations of cell membrane that appear as linear streaks [4] , as seen in Figure 4A ( green , with Cav1 . 2 in red ) . As indicated by the structures near the arrows in the right panel of Figure 4A , Cav1 . 2 localizes to exogenous BIN1 , just as Cav1 . 2 localizes to endogenous BIN1 in primary cardiomyocytes seen in Figure 1 . To confirm that membrane delivery of Cav1 . 2 to BIN1 can occur in non-myocyte cells , we evaluated surface expression patterns of exogenous Cav1 . 2 in HeLa cells expressing exogenous BIN1 . In order to resolve BIN1 structures at the level of plasma membrane , we used TIRFm , which limits the imaging depth to within 50–100 nm . Using Cav1 . 2 and BIN1 tagged with spectrally distinct fluorophores , we performed a brief time lapse capture , with representative results shown in Figure 4B . The data indicate that BIN1-induced structures ( green ) attract surface Cav1 . 2 ( red ) , causing local enrichment of calcium channel . Thereby , in the absence of other myocyte structures as well as the absence of endogenous Cav1 . 2 , ectopic expression of BIN1 is sufficient to concentrate surface Cav1 . 2 . The possibility of close biochemical association between BIN1 and Cav1 . 2 in HeLa cells is further supported by co-immunoprecipitation of V5-tagged BIN1 and Cav1 . 2 ( Figure 4B ) . In summary , we find that microtubule-based delivery of Cav1 . 2 to tubular membrane invaginations is BIN1 dependent and is independent of other myocyte-specific structures and proteins ( Figure 4C ) . To confirm that it is specifically BIN1 , and not the BIN1-induced membrane invaginations , that localizes Cav1 . 2 , we used a truncation mutant of BIN1 . Full-length BIN1 ( 1-454 aa ) has an N-terminal BAR domain followed by a coiled-coil linkage domain and a C-terminal SH3 domain ( Figure 5A ) [4] , [24] . Following precedent [4] , we created a C-terminal truncated BIN1-BAR* ( 1-282 aa ) , which retains the ability to induce membrane invagination ( the electron microscopy membrane structures are shown in Figure 5B ) . However , BIN1-BAR* loses the ability to attract endogenous Cav1 . 2 to the nascent membrane invaginations such as those in HL-1 cells ( Figure 5C ) . With full-length BIN1 ( top row ) , endogenous Cav1 . 2 is distributed along BIN1 structures . In contrast , Cav1 . 2 ( red ) has poor colocalization with BIN1 structures ( green ) in cells transfected with BIN1-BAR* ( bottom panel ) . The effect of full-length BIN1 and BIN1-BAR* on Cav1 . 2 surface targeting was further tested by a biochemical surface biotinylation assay . As in Figure 6 , unlike BIN1-BAR* , full-length BIN1 has greater surface expression of Cav1 . 2 . Thus , targeting of Cav1 . 2 to membrane invaginations requires full-length BIN1 . It appears that BIN1 recruitment of Cav1 . 2 involves a domain distinct from that which induces membrane curvature . To determine specificity of BIN1 to Cav1 . 2 , we repeated the surface biotinylation assay for the sodium calcium exchanger 1 ( NCX1 ) , which is also a T-tubule localized channel . As seen in Figure S7 , BIN1 fails to increase surface expression of NCX1 , indicating that BIN1-based delivery has specificity for Cav1 . 2 . We then investigated whether disruption of such a T-tubule-targeting mechanism of Cav1 . 2 impacts cardiomyocyte function . Although T-tubules are only partially developed in freshly dissociated neonatal cardiomyocytes [25] , [26] , earlier studies by electron microscopy show T-tubules develop after 3 days differentiation in culture [27] along with redistribution of Z-line-associated cytoskeleton proteins for Z-line organization [28] . Recent studies also find that in cultured differentiated neonatal cardiomyocytes , the dihydropyridine receptor [29] and other components of the calcium-release and uptake machinery [30] , as well as other T-tubule proteins [31] , develop a typical T-tubule staining pattern . Similarly , we observed T-tubule-type staining in cells dissociated at postnatal day three or four and allowed to differentiate in culture for a week . These structures were enriched with both Cav1 . 2 and BIN1 ( Figure S8A ) . Furthermore , BIN1 mRNA expression in postnatal mouse heart tissue is similar to that in adult heart ( Figure S8B ) . Using this differentiated mouse cardiomyocyte population , BIN1 siRNA successfully decreases BIN1 expression by 80% , as assayed by Western blot in Figure 7A . As a result of BIN1 knockdown , surface Cav1 . 2 is reduced by 45% , although the total cellular protein expression of Cav1 . 2 remains similar ( Figure 7B ) . To assay the effect on cardiomyocyte calcium transients , we loaded the cells with a fluo 4-AM and imaged with a wide-field epifluorescence microscope . As seen in Figure 7C , loss of BIN1 results in a significant slowing of calcium transient development , indicating reduced CICR . The slowing of calcium transient development is quantified by measuring the time to reach 50% of peak calcium concentration ( T1/2 max ) . BIN1 knockdown delayed T1/2 max by 40% ( bar graph , Figure 7C ) . The data in Figure 7 indicate that knockdown of BIN1 reduces the surface expression of Cav1 . 2 , impairing the intracellular cardiac transient , and that BIN1 is necessary to maintain normal calcium signaling in the heart .
The primary finding of this study is the identification of a novel role for BIN1 as a T-tubule anchoring protein accepting antegrade delivery of Cav1 . 2 . Immunocytochemical staining indicates that BIN1 colocalizes with Cav1 . 2 along T-tubules in primary adult human and mouse cardiomyocytes ( Figure 1 ) . Dual immunogold transmission electron microscopy images reveal that Cav1 . 2 and BIN1 , which co-immunoprecipitate ( Figure 4 ) , cluster together within ∼10–50 nm on T-tubules ( Figure 1C ) . Regarding delivery of Cav1 . 2 to T-tubules , there is significant support for membrane ion channel delivery occurring via microtubules [13] , [20] , [21] . In exploring the forward trafficking mechanism of Cav1 . 2 , we find that microtubules are required for the delivery of Cav1 . 2 ( Figure 3 ) and that BIN1 anchors microtubules ( Figure 2 , Table 1 ) , which can provide offloading of Cav1 . 2-containing vesicles to T-tubule membrane . BIN1 is a member of the BAR domain containing protein family , which has a role in membrane bilayer deformation at endocytic sites through interaction between their N-terminal positively charged BAR domains and acidic phospholipids within cell membrane [5] . Fluorescence and electron microscopy reveal that a human BIN1 construct can induce enormous membrane invaginations in both non-T-tubule-forming atrial HL-1 cells and non-cardiac HeLa cells ( Figures 4A , 4B , 5B ) , as previously reported in other cell types [4] . If BIN1 at cardiac T-tubules is closely associated with Cav1 . 2 ( Figure 1 ) and dynamic microtubules ( Figure 2 ) , it is possible that BIN1 alone is sufficient to target microtubule-transported Cav1 . 2 . In myocyte HL-1 cells , overexpression of exogenous BIN1 changes the cellular distribution of endogenous Cav1 . 2 and relocalizes them to nascent BIN1-induced membrane invaginations ( Figure 4A ) . Furthermore , loss of BIN1 in cardiomyocytes reduces surface expression of Cav1 . 2 ( Figure 7B ) . Such delivery is not myocyte dependent . In HeLa cells , which are devoid of the cardiac-specific cellular ultrastructures and machinery , overexpression of BIN1 is sufficient to localize exogenous Cav1 . 2 to the cell periphery on BIN1 membrane structures as resolved by simultaneous dual-color TIRFm ( Figure 4B ) . Co-immunoprecipitation of BIN1 and Cav1 . 2 ( Figure 4B ) further indicates that they are present in the same protein complex . This physical association between BIN1 and Cav1 . 2 further supports the model that BIN1 serves as a membrane anchor site for Cav1 . 2 ( Figure 4C ) . T-tubules are a well-organized membrane structure in which it is unknown how the T-tubule-related proteins localize there , specifically for Cav1 . 2 . How could we then exclude the possibility that the membrane invaginations alone are sufficient to cause Cav1 . 2 delivery to T-tubules independent of BIN1 ? As previously established [4] , the extended BAR domain ( BAR* , amino acid 1–282 , Figure 5A ) of BIN1 is sufficient for inducing membrane invagination ( Figure 5B ) , despite the absence of other domains responsible for protein-protein interaction . Distinct from full-length BIN1 , BIN1-BAR* neither redistributes Cav1 . 2 ( Figure 5C ) nor causes Cav1 . 2 surface expression ( Figure 6 ) . It appears that BIN1 recruitment of Cav1 . 2 involves a domain distinct from the one that induces membrane curvature , as suggested by endophilin binding to Cav1 . 2 in its non-BAR coiled-coil region [14] . The BAR domain superfamily has previously been associated with anchoring cortical actin at the plasma membrane [15] , [16] . This study introduces microtubule anchoring as well ( Figure 2 ) . There may be a general role for BAR domain proteins in allowing antegrade trafficking and localization of membrane-bound proteins . Furthermore , the mechanism of targeting Cav1 . 2 to T-tubules may be responsible for diseases associated with genetic BIN1 dysfunction . In mice , BIN1 knockout causes perinatal lethal cardiomyopathy [6] , and a mutation in the same 2q14-22 locus of BIN1 is associated with familial cardiomyopathy in humans [32] . Loss of function mutations in BIN1 also results in centronuclear peripheral myopathy [24] characterized by muscle weakness , which could be explained by calcium dysregulation . Future studies will be required to investigate whether calcium channel trafficking and localization are altered in these diseases . With regard to cardiac myocytes , our findings constitute a new understanding of calcium channel regulation . In order to allow trans-sarcolemmal calcium to reach the intracellular ryanodine receptors , Cav1 . 2 channels must be localized at T-tubules [9] . It has been estimated that T-tubule calcium channels contribute 80%–90% of the total cellular calcium current [10] , [11] . Furthermore , Cav1 . 2 experiences a high turnover , with pulse chase experiments indicating a half-life as short as 3 . 5 h [33] . The need for specific localization with a rapid turnover implicates that channel delivery is an important and highly regulated aspect of Cav1 . 2 channel function . Indeed , our data indicate that T-tubule targeting of Cav1 . 2 by BIN1 is critical in calcium handling and regulation in cardiomyocytes . As seen in Figure 7 , loss of BIN1 reduces surface Cav1 . 2 and delays calcium transient development in primary cardiomyocytes . The data indicate that BIN1 functions as a T-tubule-membrane-anchoring site for microtubules to deliver Cav1 . 2 , thereby ensuring proper control of cardiac EC coupling . Moreover , the mechanistic understanding of Cav1 . 2 trafficking to T-tubules by our current study not only provides insight into calcium regulation in normal hearts but also has significant implications in the pathogenesis of diseases with altered calcium dynamics such as congestive heart failure ( CHF ) . In failing heart , the intracellular calcium transient of ventricular cardiomyocytes has a low amplitude and slow decline [34]–[36] , resulting in compromised contraction [37] . Multiple factors downstream of calcium entry through Cav1 . 2 have been identified in failing muscle that contribute to changes in the calcium transient , including dysfunction in calcium removal [38] , [39] and , more recently , phosphorylation and perturbation of the ryanodine release channels [40] , [41] . There have also been reports that dyssynchronous CICR may exist in failing cardiomyocytes and contribute to defective EC-coupling gain in failing heart [42] , [43] . Since localization of L-type calcium channels is critical for synchronous CICR , loss or mislocalization of these channels in the local microenvironment might lead to defective CICR and abnormal heart function . In fact , human CHF has reduced L-type calcium channel density in the sarcolemma [44] , and a canine model of heart failure is associated with remodeling of both Cav1 . 2 distribution and T-tubule structure [45] . In this study , we found that BIN1-based microtubule targeting affects Cav1 . 2 localization and intracellular calcium dynamics ( Figure 7 ) . It will be interesting in future studies to explore the role of BIN1 regulation in heart failure .
Human BIN1 ( Isotype 8 ) cDNA was obtained from Origene . Full-length BIN1-8 ( 1-454 aa ) and BIN1-BAR* ( 1-282 aa ) were then amplified and cloned into pDONR/Zeo ( Invitrogen ) using Gateway BP cloning to generate entry clones . The genes were subsequently inserted into pDest-eGFP-N1 , pDest-mCherry-N1 ( converted vectors originally from Clontech ) , and pcDNA3 . 2-V5-Dest by Gateway LR cloning . Human Cav1 . 2 was obtained from Origene . Human β2b and rabbit α2δ1 were generously provided by Dr . Michael Sangunetti . N-terminal GFP-Cav1 . 2 was generously provided by Dr . Kurt Beam , and C-terminal Cav1 . 2-GFP was described previously [46] . Non-targeting and BIN1-specific siRNA were obtained from Dharmacon . HeLa cells and mouse atrial HL-1 cells were cultured in DMEM and Claycomb medium under standard mammalian cell conditions . FuGene 6 ( Roche ) was used for cDNA transfections in HeLa cells . Lipofectamine ( Invitrogen ) was used for cDNA transfections in HL-1 cells . Dissociated cardiomyocytes were allowed to attach to laminin-precoated glass coverslips before fixation . For all immunocytochemistry , cells were fixed in methanol at −20°C for 5 min . For immunohistochemistry , cryosections were fixed in ice-cold acetone for 10 min . After fixation , cardiomyocytes were permeablized and blocked with 0 . 5% Triton X-100 ( Sigma ) and 5% NGS in PBS for 1 h at room temperature . For BIN1 and Cav1 . 2 staining , the cells were incubated with mouse anti-BIN1 ( 1∶50 , Sigma ) and rabbit anti-Cav1 . 2 ( 1∶50 , Alomone ) overnight at 4°C . Similar protocol without permeablization was used for BIN1 and Cav1 . 2 staining in myocardium cryosections . For co-staining of Cav1 . 2 and α-tubulin in HL-1 cells , after fixation , the cells were permeablized with 0 . 1% Triton X-100 for 15 min and blocked with 5% NGS for 1 h . The cells were then incubated with rabbit anti-Cav1 . 2 ( 1∶50 , Alomone ) overnight at 4°C followed by mouse monoclonal to α-Tubulin ( 1∶500 , Sigma ) for 1 h at room temperature . After several washes with PBS post-primary antibody incubation , cells were then incubated with goat anti-mouse and -rabbit IgG conjugated to AlexaFluor 488 and 555 , respectively . Cells were then fixed and mounted with DAPI containing ProLong gold . All imaging was performed on a Nikon Eclipse Ti microscope with a 100× 1 . 49 NA TIRF objective and NIS Elements software . Deconvolution of images was performed using Autoquant software ( Media Cybernetics ) . High-resolution cardiomyocyte images were obtained by a spinning disc confocal unit ( Yokogawa CSU10 ) with DPSS lasers ( 486 , 561 ) generated from laser merge module 5 ( Spectral applied research , CA ) and captured by a high-resolution Cool SNAP HQ2 camera ( Photometrics ) . Multiple wavelength TIRF was achieved with Dual-View emission splitter ( Optical Insights ) . High-sensitive Cascade II 512 camera ( Photometrics ) was used for TIRF image capture . For BIN1 and Cav1 . 2 distribution , isolated human and mouse cardiomyocytes were imaged at Z-depth increments of 0 . 1 µm and reconstructed to generate three-dimensional volume views and frame view along the longitudinal axis using NIS Element Software . To access Cav1 . 2 and BIN1 colocalization by TIRFm , HeLa cells were plated overnight and co-transfected with pDest-BIN1-mCherry , and Cav1 . 2-GFP along with β2b and a2δ1 . Dual channel TIRF time lapse sequences of 1 min were acquired at an exposure of 200 ms per image at a rate of 1 frame per second . After acquisition , the total 61 frames were z-projected into one frame using ImageJ ( NIH ) . For live-cell imaging of microtubule behavior by spinning disc confocal microscopy , Hela cells were plated and co-transfected with pDest-BIN1-mCherry and α-tubulin-GFP . Time lapse sequence for α-tubulin was acquired at a continuous rate of 1 s with 400 ms exposure per frame . To confirm the similar BIN1 expression pattern , BIN1 images were taken both before and after tubulin time lapse sequence . The tubulin-GFP particle paths were manually traced and analyzed for travel velocity and pause event in the time sequence using MTrackJ Plugin in ImageJ . “Microtubule at BIN1” is considered when the microtubule tip is within 0 . 2 µm of the closet BIN1 edge . For calcium imaging in neonatal mouse cardiomyocytes , cardiomyocytes were loaded with a cell permeable calcium dye 4-AM in calcium-free HBSS ( Gibco ) for 15 min and imaged in regular HBSS ( Gibco ) with a 20× TIRF objective with a wide-field epifluorescence microscopy . Live images were captured by a high-sensitive Cascade II 512 camera at a frame rate of 64 ms for 20 s . For electron microscopy membrane ultrastructure , cells were fixed in Karnovsky's fixative ( 1% paraformaldehyde / 3% Glutaraldehyde in 0 . 1 M Sodium cacodylate buffer , pH 7 . 4 ) at room temperature for 30 min before being stored at 4°C . The method for the membrane ultrastructure study was previously described [47] , [48] . Briefly , the fixed cells were then post-fixed in OsO4 ( 2% OsO4 + 1 . 5% potassium ferrocyanide , Sigma ) and stained en bloc with 1% tannic acid ( Sigma ) , uranyl acetate ( EM Science ) before being dehydrated in ethanol , cleared in propyline oxide , and embedded in eponate 12 ( Ted Pella Co . ) . Finally , cells were sectioned and stained with uranyl acetate and Reynold's Lead to enhance contrast and were examined under Philips Tecnai 10 electron microscope ( Eidhoven ) . For immunolabeling , mouse cardiomyocyte suspension was fixed in 2% paraformaldehyde / 0 . 1% glutaraldehyde in 0 . 1 M cacodylate buffer pH7 . 4 at room temperature for ∼2–3 h . An established procedure [49] , [50] was used for immunogold labeling of mouse cardiomyocytes . Briefly , the fixed samples were cryoprotected with PVP/sucrose ( 20% polyvinyl pyrrolidone [Sigma] in 2 . 3 M sucrose ) overnight and frozen in liquid nitrogen before being cut into thin sections with Leica Ultracut UCT with EMFCS attachment ( Leica Microsystems Inc . ) . Sections were treated with 0 . 2% glycine , blocked with 2% BSA/gelatin in PBS , pH 7 . 4 , incubated with mouse anti-BIN1 ( 1∶2 , Sigma ) and rabbit anti-Cav1 . 2 ( 1∶2 , Alomone ) diluted with blocking solution overnight at room temperature ( controls were done with normal mouse serum ) , and incubated with 10 nm immunogold conjugated anti-mouse ( 1∶25 ) and 15 nm immunogold conjugated goat anti-rabbit ( 1∶50 ) secondary antibodies for 30 min . The sections were then stained with oxalate uranyl acetate and embedded in 1 . 5% methyl cellulose ( Sigma ) and 0 . 3% aqueous uranyl acetate ( Ted Pella Inc . ) . Colocalization between Cav1 . 2 and BIN1 was examined in a Philips Tecnai 10 electron microscope ( Eidhoven ) . With the approval of the University of California–San Francisco ( UCSF ) Committee for Human Research , we obtained tissue from organ donors whose hearts were not transplanted . The California Transplant Donor Network ( CTDN ) provided the unused donor hearts and obtained informed consent for their use from the next of kin . All the mouse work was approved by UCSF Committee for Animal Research . All procedures were in accordance with UCSF animal research and care protocols . After immediate perfusion with cold cardioplegia , full-thickness samples from left ventricular free wall were cleaned rapidly of all epicardial fat and snap frozen into liquid nitrogen for later protein and mRNA analysis . More sections were embedded in OCT medium and frozen in liquid N2-chilled isopentane for immunohistochemistry . For cardiomyocytes isolation , ventricular free wall samples were cut into ∼1 mm3 sections for digestion with pre-warmed collagenase II ( 2 mg/ml , Worthington ) at 37°C in calcium-free KHB solution ( 134 mM NaCl , 11 mM Glucose , 10 mM Hepes , 4 mM KCl , 1 . 2 mM MgSO4 , 1 . 2 mM Na2HPO4 , 10 mM BDM , 0 . 5 mg/ml BSA , Ph 7 . 4 ) [51] with modification of a previously reported method [52] . Dissociated cardiomyocytes were allowed to attach to laminin-precoated glass coverslips before fixation for immunocytochemistry . Mouse ventricular myocytes were isolated from male adult C6/Black mouse ( ∼8–12 wk; Charles River ) after dissociation with collagenase II ( 2 mg/ml , Worthington ) with a previously described method [53] . For surface biotinylation experiments , cardiomyocytes were attached to laminin-precoated culture dishes and cultured in primary cardiomyocyte medium ( ScienCell ) in 37°C and 5% CO2 incubator . The cells were treated with vehicle ( DMSO , 1∶2000 ) overnight ( 16 h ) before the replacement with control medium ( containing DMSO , 1∶2000 ) or medium containing 20 µM dynasore with or without 30 µM nocodazole for 2 h . For 18 h nocodazole treatment , cardiomyocytes were cultured in medium containing 30 µM nocodazole overnight ( 16 h ) before the replacement of medium containing dynasore + nocodazole for another 2 h . Timed pregnant mice were ordered from Charles River at E16-17 . Primary mouse neonatal cardiomyocytes were isolated from p3/4 C57BL/6 mice and maintained in F12/DMEM 50/50 ( Invitrogen ) supplemented with 2% FBS , Insulin-transferrin-sodium selenite media supplement , 10 µM 5-Bromo-2′-deocyuridine , 20 µM Cytosine β-D-arabinofuranoside ( Sigma ) , and 100 µg/ml Primocin ( Amaxa ) . Cells were maintained in a humidified atmosphere of 5% CO2 at 37°C . Cardiomyocytes were allowed for differentiation in culture for about a week before surface biotinylation and calcium-imaging experiments . After 3 to 4 d in culture , the cells were transfected with 125 nM control or BIN1 siRNA ( Dharmacon ) , which was repeated 24 h later . Three days after the first dose of siRNA , surface biotinylation experiments and calcium imaging were studied in these cells . After treatment , the cells were quickly washed and incubated with ice-cold 1 mg/ml High Capacity Neutraavidin Agarose Resin ( Pierce ) for 25 min . After 2 × 5 min quenching of unbound biotin with 100 µM glycine , cells were washed and lysed in RIPA buffer ( 50 mM Tris pH 7 . 4 , 150 mM NaCl , 1 mM EDTA , 1% Triton X-100 , 1% sodium deoxycholate , 2 mM NaF , 200 µM Na3VO4 ) supplemented with Complete Mini protease inhibitor cocktail ( Roche ) . Total protein concentrations were determined and normalized between samples . The lysates were then incubated with prewashed NeutrAvidin coated beads at 4°C overnight . After washes , bound surface proteins were eluted and boiled , separated on NuPage gels ( Invitrogen ) , and probed with rabbit anti-Cav1 . 2 antibody ( Alomone ) and mouse anti-NCX1 antibody ( Abcam ) . Similar expression levels of BIN1 and BIN1-BAR* were confirmed by Western blot in the total cellular lysates . For quantitation , the amount of surface Cav1 . 2 or NCX1 was normalized to input and compared among different groups . HeLa cells were cotransfected with human Cav1 . 2 along with regulatory β2b and α2δ1 subunits and BIN1-V5 , harvested , and lysed in 1% Triton X-100 Co-IP buffer ( 50 mM Tris pH 7 . 5 , 150 mM NaCl , 2 mM EDTA , 2 mM EGTA , 1 mM DTT , 1 mM NaF , 100 µM Na3VO4 , 1% Triton X-100 ) supplemented with Complete Mini protease inhibitor cocktail . The lysate was then incubated with either mouse anti-V5 antibody ( 2 µg ) or equal amount of non-specific mouse IgG for 2 h before pulldown with rec-protein-G-Sepharose ( Invitrogen ) for 1 h . Material bound to washed beads was eluted , boiled , separated , and probed with rabbit antibodies against Cav1 . 2 ( Alomone ) or V5 ( Sigma ) . For spatial periodicity analysis in the cardiomyocytes , the fluorescence intensity profiles were generated by ImageJ . The frequency domain power spectrum of cardiomyocyte subsections were generated in Matlab using FFT conversion . Next , the power spectrum over spatial distance ( 1/frequency ) was averaged from five cardiomyocytes and presented in Figure 1 . For T-tubule Cav1 . 2 signal , intensity at each peak ( corresponding to T-tubules ) was analyzed using the fluorescence intensity profiles generated by ImageJ and Matlab . For quantitation of cell peripheral Cav1 . 2 , three-dimensional cross-section projection of cardiomyocytes were generated , and fluorescence intensity within 2 µm of cell surface was analyzed using ImageJ . In addition , a previously reported method [54] using PSC Colocalization plug-in in ImageJ was used for colocalization analysis between BIN1 and Cav1 . 2 . For all other statistical analysis , paired or unpaired two-tail Student's t test was performed using Prism 5 ( GraphPad ) software . | Calcium plays a primary role in regulating heart function . During each heartbeat , calcium ions cross the membrane of individual cardiac muscle cells and trigger a rapid increase of calcium within the cell ( called the calcium transient ) . Calcium causes the muscle cells to contract and determines the strength of the overall heartbeat . Each cardiac muscle cell has many small tubular-like membrane invaginations known as T-tubules where calcium channels localize , allowing calcium ions to enter and immediately encounter intracellular calcium release organelles . While this organization is well described , it is not known how calcium channels localize to T-tubule membrane . Here we show that in human and mouse heart cells , a membrane scaffolding protein known as BIN1 is localized together with calcium channels at T-tubules . Using high-resolution live cell microscopy , we found that microtubules , which are necessary for calcium channel delivery to the membrane , are also tethered by BIN1 . Loss of BIN1 in cardiac cells impairs delivery of calcium channels to the membrane and diminishes the intracellular calcium transient . According to this model , microtubules function as highways that carry newly synthesized calcium channels to BIN1-containing membrane . Once tethered to T-tubules by BIN1 , the microtubules can deliver their calcium channel cargo . We postulate that this calcium channel delivery pathway is important to the regulation of cardiac calcium signaling and beat-to-beat cardiac function . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"cell",
"biology/membranes",
"and",
"sorting",
"cardiovascular",
"disorders/myopathies",
"cardiovascular",
"disorders/heart",
"failure",
"cell",
"biology/cytoskeleton"
] | 2010 | BIN1 Localizes the L-Type Calcium Channel to Cardiac T-Tubules |
Evidence from human genetic studies of several disorders suggests that interactions between alleles at multiple genes play an important role in influencing phenotypic expression . Analytical methods for identifying Mendelian disease genes are not appropriate when applied to common multigenic diseases , because such methods investigate association with the phenotype only one genetic locus at a time . New strategies are needed that can capture the spectrum of genetic effects , from Mendelian to multifactorial epistasis . Random Forests ( RF ) and Relief-F are two powerful machine-learning methods that have been studied as filters for genetic case-control data due to their ability to account for the context of alleles at multiple genes when scoring the relevance of individual genetic variants to the phenotype . However , when variants interact strongly , the independence assumption of RF in the tree node-splitting criterion leads to diminished importance scores for relevant variants . Relief-F , on the other hand , was designed to detect strong interactions but is sensitive to large backgrounds of variants that are irrelevant to classification of the phenotype , which is an acute problem in genome-wide association studies . To overcome the weaknesses of these data mining approaches , we develop Evaporative Cooling ( EC ) feature selection , a flexible machine learning method that can integrate multiple importance scores while removing irrelevant genetic variants . To characterize detailed interactions , we construct a genetic-association interaction network ( GAIN ) , whose edges quantify the synergy between variants with respect to the phenotype . We use simulation analysis to show that EC is able to identify a wide range of interaction effects in genetic association data . We apply the EC filter to a smallpox vaccine cohort study of single nucleotide polymorphisms ( SNPs ) and infer a GAIN for a collection of SNPs associated with adverse events . Our results suggest an important role for hubs in SNP disease susceptibility networks . The software is available at http://sites . google . com/site/McKinneyLab/software .
Human genetics studies have been successful at identifying single-locus variants that have a large effect on Mendelian disorders , such as cystic fibrosis or neurofibromatosis . However , the analytical strategies appropriate for identifying Mendelian disease genes have been met with limited success when applied to common multigenic diseases [1] , [2] . Contributing to this limited success is the fact that the Mendelian approach requires that each susceptibility factor exert a large independent ( main ) effect on disease risk because association with the phenotype is investigated only one genetic locus at a time . The complexity of molecular interactions necessary to regulate gene expression likely is reflected at the DNA sequence level in the form of statistical interactions between alleles , with many of the individual alleles having little or no main effect on disease risk . The breakdown in the buffering against complex disease-related changes in expression may only be observable if properly investigated in terms of statistical interactions between genetic variants like single nucleotide polymorphisms ( SNPs ) or copy number polymorphisms ( CNPs ) . Thus , analytical strategies for genetic association studies are needed that identify conditionally-dependent ( interacting ) susceptibility factors in addition to factors that exhibit an independent effect . Such strategies must be able to capture the spectrum of Mendelian to multifactor interaction effects . Gene–gene interaction is widely accepted in the field of statistical genetics as a significant challenge to understanding the genetic architecture of complex diseases [3]–[9] . There is empirical evidence from human studies and model organisms to suggest that gene–gene interactions contribute to variation in complex diseases [10]–[15] . In human studies , for example , interactions were detected in Alzheimer disease between GAB2 and APOE [16] , and high-risk haplotypes displaying intralocus interactions were detected in exfoliation glaucoma [17] and atrial fibrillation [18] . Another notable example of the importance of interactions in human disease is Hirschsprung's disease which was found to be influenced by polymorphisms in RET and EDNRB in the Old Order Amish [19] . This association was confirmed in a mouse model and the synergistic effect of both variants greatly outweighed the additive risk of each variant when considered independently . Like the examples above , susceptibility to common diseases , such as cancer , diabetes , obesity , hypertension , and premature cardiovascular disease , is likely influenced by the interaction of SNPs in multiple genes . Moreover , Mendelian disorders display a wide range of phenotypic variation that may be explained by interactions of the primary mutation with genetic modifier variants . Our working definition of gene–gene interaction is the conditional dependence between genetic variants that affects the classification of the phenotype . This definition is equivalent to the definition based on deviation from additivity in a multi-locus model of phenotypic variation . These interactions may vary from very weak , or nearly additive , to purely epistasic , where it is only possible to detect a susceptibility locus when considered jointly with one or more additional loci . An advantage of genome-wide association ( GWA ) studies is that information about conditionally dependent loci is more likely to be available for gene–gene interaction analysis . Unfortunately , these useful genotypes are embedded in a genome-wide sea of noise , or variants irrelevant to classification of the phenotype . Thus , the focus of this paper is to address these two challenges in GWA studies: 1 ) accounting for gene–gene interactions and main effects and 2 ) removing noise variants to obtain a subset of SNPs that are enriched for association with the phenotype . Random Forests ( RF ) [20] is a powerful nonparametric method that has been successfully applied to genetic data [21] . An RF is a collection of decision tree classifiers in which each tree in the forest has been trained on a bootstrap sample of instances from the data and each split attribute is chosen from among a random subset of attributes . In data mining terminology , an attribute is a dataset feature or variant such as a SNP , and an instance refers to a sample or subject . Out-of-bag instances are used to estimate prediction error and importance of each attribute via permutation testing . If randomly permuting values of a particular attribute does not affect the predictive ability of trees on out-of-bag samples , then that attribute will be assigned a low importance score [5] , [21] . RF has been targeted as a method for identifying interactions in genetic data because it takes into account the context of other attributes when scoring the relevance of individual genetic variants and it does not require the specification of a model [22] . However , when association of an attribute with the phenotype is caused by a pure interaction with another attribute , the RF importance score of the relevant attribute diminishes . This limited ability to identify interacting attributes is due to the independence assumption used during node splitting , which is determined by the Gini index . The resulting trees are built on the assumed independence of the split attribute conditional on the class because the Gini split selector measures the impurity of the class value ( case or control ) distribution before and after the split on the evaluated attribute ( e . g . , SNP ) . Recursive Elimination of Features-F ( Relief-F ) is a heuristic attribute quality measure that can identify important variants in data sets that include strong interactions . However , Relief-F is sensitive to the presence of noise attributes , which when added to the data set cause Relief-F scores of relevant variants to worsen [23] . This limitation is exacerbated in GWA studies in which most of the variants may be irrelevant to the given phenotype . To overcome the bias caused by the context of noise attributes , strategies are necessary that iteratively remove variables with the worst Relief-F scores and update the scores of the remaining variables [23] , [24] . The authors in Ref . [24] applied such a strategy , called tuned Relief-F ( TuRF ) , to simulated genetic association data and demonstrated increased power to identify interacting SNPs over Relief-F without backwards elimination . Recently , we used evaporative cooling ( EC ) to create a composite score from Relief-F and information gain ( IG ) , thereby demonstrating greater power than iterative Relief-F to detect pure interactions , with markedly greater power observed when one of the interaction partners demonstrated a marginal main effect . In real genome-wide association data , one expects both interaction and main effects to be present . Thus , the motivation of the EC filter is to optimize the linear combination of complementary scores to detect the continuum of independent and interaction effects . The EC approach in the current study optimizes the coupling of the RF and Relief-F scores based on classification accuracy and the iterative removal of noise attributes to obtain a collection of SNPs enriched for relevance to the phenotype . The development of EC as a machine learning method was motivated by information theory and the statistical thermodynamics of cooling a gas of atoms by evaporation [26] . Just as a balance is struck between low energy and high entropy to achieve equilibrium in a collection of atoms , EC feature selection balances independent and interaction effects to obtain a collection of attributes enriched for association with the phenotype . Further , EC of a physical gas increases the phase space density by the iterative removal of the most energetic atoms while EC feature selection increases the feature space density by iteratively removing attributes that are least relevant to the phenotype . In a physical system , energy ( E ) and entropy ( S ) are balanced through the free energy F = E−TS , where T is the system temperature . In EC feature selection we optimize an analogous quantity that we call the information free energy , where E is the interaction contribution ( Relief-F ) and S is the main effect contribution . These two quantities are balanced by optimizing the coupling T . In our previous construction of the information free energy score we used IG , a quantity derived from information entropy , because the S contribution represents entropy in the thermodynamic free energy [26] . However , EC is not restricted to rely on an information entropy-based correlation score and , in fact , EC has a flexibility that allows it to couple any attribute quality scores . Thus , for the current study we use Relief-F as the interaction score and a transformation of the RF importance score as the independent effect score . Through simulation analysis we show that the EC filter is able to identify genetic variants that confer risk through interaction with other genetic factors . Such risk factors may go undetected in a typical GWA analysis that reports a stringent list of the most significant SNPs where each SNP has been treated as independent . We apply the EC interaction filter to a real data set , which we analyzed previously for main effects using logistic regression ( LR ) [27] . The data set consists of 1442 SNPs across 386 candidate genes for subjects with and without systemic adverse events following smallpox vaccination . In order to characterize the interactions among the top EC-ranked SNPs , we infer what we call a genetic-association interaction network ( GAIN ) . GAIN is based on interaction information ( II ) , which was formulated by McGill [28] to quantify higher-order interaction gains between attributes and the class or phenotype . Jakulin and Bratko in Ref . [29] proposed a number of novel diagrams to visualize these interactions , some of which were incorporated by [30] into a strategy to characterize epistasis in multifactor dimensionality reduction ( MDR ) models . Positive connection strength between SNPs in a GAIN represents synergy between the two SNPs whose joint variation leads to improved classification of the phenotype . A negative network connection indicates redundant information between the two SNPs . In the terminology of genetics , “synergy” maps onto epistasis , and “redundancy” is most closely related to linkage disequilibrium but conditional on the phenotype . The EC filter , with its ability to select SNPs that may involve interactions or main effects , combined with GAIN for visualization and interpretation of the resulting network , provides an alternative approach to analyzing genotypic data on a more global scale , which will become increasingly important as GWA studies become more prevalent .
Figure 1 depicts the two-locus interaction models simulated in this study to compare the performance of EC , TuRF , RF , and stepwise penalized LR ( stepPLR ) [31] . The models in Figure 1 include combinations of low heritabilities ( h2 = . 05 on the left and . 01 on the right ) and a range of interaction strengths ( from nearly additive to completely epistatic ) . The 1% level represents a worst-case scenario for heritability , and the purely epistatic XOR model ( Model 3 ) represents the worst-case scenario for gene–gene interaction models . For each genetic model , 100 replicate datasets were created with 1000 samples consisting of a balanced number of cases and controls . The proportions of the susceptibility alleles A and B in the population are assumed to be the same as the alleles a and b , respectively . Replicate simulations were created using the genomeSim software [32] . In addition to simulating the specified interaction model , genomeSim also simulates linkage disequilibrium ( LD ) patterns , though LD is not the focus of the current study . Each replicate dataset consists of a set of 1500 SNPs containing the two susceptibility SNPs . Figure 2 summarizes the comparison of the ability of EC , TuRF , RF , and stepPLR to detect two-locus models described in Figure 1 . For the 100 replicates of each model in Figure 1 , we recorded the number of times that the two susceptibility SNPs were detected among the top filtered SNPs for each analytical method . The empirical detection power in Figure 2 is defined as the fraction of times out of all 100 replicate data sets for a given model that both of the simulated susceptibility SNPs occurred in the top SNPs as ranked by the given method . The cutoff for how many SNPs to include from the rank-list is varied in Figure 2 from the top 2 to the top 100 SNPs . In a real data analysis one may choose a filter cutoff that is larger than the top 2 because findings that replicate in multistage study designs often are not the most statistically significant associations in the initial scan [33] , [34] . And , as we show below , a larger collection of SNPs permits a pathway-level analysis in which SNPs in multiple genes contribute to disease risk . Below we illustrate on a real data set a random permutation approach for selecting a significant EC cutoff score . When determining the top SNPs for EC , RF and TuRF , we sort by importance score . Detection is counted for stepPLR if both causal SNPs have a nonzero coefficient anywhere in the LR model . For RF , we used 10 , 000 trees in a forest and the square root of the total number of SNPs as the number of SNPs chosen randomly for node splitting . We used 10 nearest neighbors in the Relief-F calculations . For both RF and Relief-F , we used iterative removal of irrelevant SNPs in order to compare with EC consistently . The simulation results ( Figure 2 ) show that EC does as well as , or improves upon , RF for all interaction models . For the interaction models with a small main effect ( Models 1 , 2 , and 4 ) , EC and RF perform similarly . For the interaction Model 2 with a small main effect and low ( 1% ) heritability , EC and RF display modest power in the 20–25% range , while the power of TuRF is even lower at 7% . The weakness of RF at identifying purely epistatic models is most evident for Model 3 , which has a relatively high ( 5% ) heritability but is an XOR model with zero marginal effect . When the final number of SNPs is two , RF has only a 14% detection power for Model 3 whereas EC detects it with 91% power . For the interaction Model 2 with a non-vanishing main effect , all methods perform poorly due to the low ( 1% ) heritability . It is likely that any analytical method would need a larger sample size to detect Model 2 with appreciable power . StepPLR shows a constant power for all models because of the small regularization parameter chosen by cross validation , which leads to models with fewer variables . For Model 4 , StepPLR has an advantage over other methods when restricted to choosing the top two SNPs . EC performs better than TuRF for all models tested . TuRF shows its best performance for the XOR model , but performs worse than StepPLR for Models 1 and 4 , which have an additive effect . By combining RF and Relief-F , the EC algorithm is able to detect interaction models with slight main effects , for which RF is well suited , and , leveraging the strength of Relief-F , EC is able to detect pure interaction models that RF is too myopic to detect . To illustrate the potential for detecting larger genetic models , Figure 3 shows the results of an analysis of 100 simulated replicates of an 8-locus model that combines the two-locus models from Figure 1 . We compare the detection frequency for each of the eight functional loci for EC with a cutoff of 50 and StepPLR . This analysis shows the potential advantage of using EC as a filter for genetic models involving greater than two loci . To further validate and illustrate our approach , we apply the EC filter and GAIN strategy to a smallpox vaccine study looking at the association of a panel of SNPs with mild adverse events ( AEs ) [27] . Of the 131 subjects in the study , 40 experienced a systemic AE , which included fever , generalized rash and lymphadenopathy . Table 1 shows the top 26 SNPs out of the 1442 ranked by the EC filter for the vaccine AE phenotype . We arrived at this cutoff using the random permutation approach described in the Methods section . An EC score cutoff of −0 . 237 yields a . 05 risk for a SNP that is declared relevant to the phenotype in Table 1 is actually irrelevant . As we dissect in more detail below , glycogen synthase kinase 3 beta ( GSK3B ) and solute carrier family 6 ( neurotransmitter transporter , dopamine ) , member 3 ( SLC6A3 ) in Table 1 are likely information hubs in this phenotype network . In addition to potential interaction effects , the EC relevance list also contains the same SNPs found in our previous main effect analysis in the 5 , 10-methylenetetrahydrofolate reductase ( MTHFR ) and interleuking-4 ( IL4 ) genes . To characterize the details of the interaction network of this SNP-phenotype network , we draw upon the top 100 EC-ranked SNPs ( see Supplementary Table 1 ) , which we reduce to 70 SNPs by removing redundant markers . Specifically , if a pair of SNPs is correlated by more than . 8 by the symmetric mutual information measure [26] , then the least informative marker ( lower EC score ) is removed . The purpose of removing correlated SNPs is to reduce redundancy and make the network more interpretable . Another way to simplify the network would be to use nodes corresponding to constructed haplotypes . We also removed the least informative marker between pairs that are redundant in the context of the phenotype according to II . For example , GSK3B_01 and GSK3B_27 are correlated by less than . 8; however , in the context of the phenotype they have nearly the same information content . Another way we plan to deal with correlated features in the future is to wrap an orthogonalization procedure into the EC method . Figure 4 shows the GAIN inferred from this EC-filtered list of SNPs . The sample size of this vaccine trial is relatively small for identifying high-order interactions; thus , our goal in Figure 4 is simply to illustrate how the EC filter may be used in conjunction with GAIN to add insight into the network of interactions among SNPs that may influence the phenotype for a typical genetic association study . Edges represent interaction information ( II ) between SNPs , which does not simply represent correlation between SNPs but rather quantifies the amount by which their joint variation decreases our uncertainty about the phenotype over what would be expected by their individual effects ( see Methods ) . For clarity of the graph , the number of connections displayed is limited to pairs of SNPs with the largest II magnitudes and to pairs involving nodes with the best EC scores . Specifically , we displayed edges between pairs of SNPs with an II magnitude greater than 6% , which results in 160 edges . The 6% II cutoff yields a . 03 risk of obtaining a false connection , which was calculated by random permutation of SNP pairs ( see Methods ) . Pairs of SNPs with positive II ( synergy between the SNPs with respect to the phenotype ) have solid edges . Pairs of SNPs with negative II ( redundant information with respect to the phenotype ) are indicated by dashed edges . We highlight three nodes and their connections in Figure 4: a hub SNP in GSK3B ( orange ) and two main effect SNPs in MTHFR ( yellow ) and IL4 ( green ) , which displayed main effect haplotypes in our primary analysis . GSK3B has a relatively large IG ( see Supplementary Table 2 for numeric details of the GAIN interaction partners ) but it may be more important in its influence on other SNPs in the context of the AE phenotype . For example , GSK3B has a direct connection with IL4 and a secondary connection with MTHFR . In the top ranking interaction pair ( Suppl . Table 2 ) , a SNP in GSK3B and lipase , hepatic ( LIPC ) contribute the most total gain in information about the phenotype despite the very small IG of LIPC . Supplementary Table 3 gives the connectivity distribution for the GAIN nodes . A SNP in SLC6A3 and GSK3B are hubs in the GAIN . Its synergy between several interaction partners and its independent effect give GSK3B one of the best EC scores ( Table 1 ) . The SNP in SLC6A3 has a much smaller main effect than GSK3B but its synergy leads to its having one of the better EC scores . Despite having a much smaller main effect , the SLC6A3 hub SNP appears to have an influential , though indirect , effect on the phenoytpe .
As pointed out by [35] an important challenge facing statistical genetics will be to balance the relative strengths and weaknesses of new and existing analytical methods because of the multiple challenges a single method must adequately address in a data set , including heterogeneity , gene–gene/gene-environment interactions , and genome-wide noise . The EC method attempts to achieve this balance via a machine learning optimization analogous to the way a system of particles achieves equilibrium by balancing low energy and high entropy as expressed through the thermodynamic free energy . The goal of EC is to address the challenge of disease model heterogeneity , gene–gene interactions , and noise by optimizing the coupling between two powerful machine learning/data mining methods with complementary strengths and weaknesses: Random Forest ( RF ) and Relief-F . The iterative removal of attributes ( evaporation ) plays a dual role by providing a mechanism for optimizing the coupling of RF and Relief-F and by removing attributes that are irrelevant to the phenotype ( noise ) from high-dimensional genotype data . Relief-F was designed to account for interacting variants but consequently is more sensitive to noise . RF is more limited in its ability to identify interaction effects but is robust to noise , overfitting , and missing data . In addition , tree-based methods are suited to deal with certain types of genetic heterogeneity because splits near the root node define separate population subsets in the data . These methods exhibit complementary strengths and weaknesses . Thus , properly integrating these two scores and using backwards elimination allows EC to identify a spectrum of interaction effects , from purely epistatic XOR models to models displaying Mendelian effects . The selection from a random sample of attributes allows RF to maintain a low correlation between trees while the coupling with Relief-F by EC enriches the population of attributes for interaction effects that influence the phenotype . Many SNPs in association studies have been shown to have small individual effect sizes , but their combined effect may be much larger . EC has high power to filter a large set of SNPs down to a small subset that is enriched for interaction and independent effects that influence association with the phenotype . The advantage of EC over standard statistical analysis is greatest when the genetic model contains no marginal main effect; however , EC performs as well as or better than RF for the interaction models that contain a main effect . EC also outperforms iterative Relief-F procedures ( e . g . , TuRF ) with the greatest improvement occurring when one of the attributes demonstrates a small main effect . Thus , by balancing independent and interaction effects , EC is able to detect a spectrum of models in genetic association studies . Currently EC is meant to be an attribute filter for dimensionality reduction to be followed by more fine-grained modeling and/or a second stage of genotyping . However , in our simulation analysis we often find that the two functional SNPs have the highest EC rank , not just in a top percentile . In the application to real data we used a permutation approach to estimate an appropriate critical region of EC scores to reduce the number of false positive SNPs . To characterize the genotype to phenotype map , we used the EC filter to reduce the space of SNPs to a more computationally manageable size for an exhaustive search for interactions by II , and then we used a network approach to visualize the interactions on a larger scale . Instead of using information theory to infer specific gene–gene interactions , one may use an alternative like pair-wise LR . The EC filter plus GAIN approach may prove to be a valuable complement to other approaches for modeling complex diseases because the inferred disease-specific network may better approximate the interconnectivity of the true biological system . In our second stage of analysis of the real data set , we inferred a genetic association interaction network ( GAIN ) in Figure 4 , by taking advantage of the context dependence of all SNPs in the EC filter score . We again used a permutation strategy to prune the network by estimating the II cutoff appropriate for the given data . GAIN provides a visualization tool to explore , more globally , the statistical and biological relationships among the SNPs that are relevant to a given phenotype . GAIN is meant to be a discovery tool to suggest a SNP interaction network of the given phenotype . It provides information about synergy between SNPs whose joint effect increases information about the phenotype as well as information about redundancy between SNPs whose joint effect provides no additional information about the phenotype over their independent contributions . The SNP hubs GSK3B , LIPC , and SLC6A3 ( Supplementary Table 3 ) in Figure 4 have some of the highest EC scores and yield the largest information gains when joined with SNPs in other genes ( Supplementary Table 2 ) . Supplementary Table 2 only shows pair-wise interactions; however , GAIN Figure 4 suggests higher-order effects that can be decomposed into pair-wise interactions that cascade from the hub . The cascading effect of such hubs on disease susceptibility is an important area of investigation as is the identification of sub-networks in the GAIN that may suggest new pathways involving the given phenotype . As EC and GAIN are further developed it will be important to integrate gene ontology ( GO ) information with GAIN so that significantly enriched GO terms can be highlighted in network motifs . We used permutation to set the number of edges , but the use of prior knowledge may also help to determine the appropriate II cutoff magnitude for displaying GAIN edges , thereby reducing the number of false positive interactions in an inferred network . For speed of analysis for large numbers of simulated data sets , this paper focused on candidate gene data sets on the order of 103 SNPs but not high-density , whole genome data , which are typically on the order of 105 or 106 SNPs . To make whole-genome filtering feasible , we have implemented a version of EC that is parallelized ( pEC ) . The freely available software for pEC results in a decrease in CPU time by a factor approximately equal to the number of processors used . Our strategy involves parallelizing the attribute quality evaluations ( RF and Relief-F ) in the evaporation loop since this is the most time consuming step of EC . We use a parallelized version of RF in Fortran 90 , parallel RF ( PARF ) [36] . Using test data sets with a sample size of 1000 cases and 1000 controls , we estimate the computational speed of the current version of EC is 1 . 5 seconds per marker . Based on this rate , EC would be able to filter 1 million SNPs in 42 hours on 10 processors . The other computational advantage of EC over exhaustive search strategies is that EC takes into account the context of all SNPs when scoring a SNP , allowing for the inclusion of higher-order effects at no additional computational cost . Despite using a Naïve Bayes classifier to determine the parameter , T , for coupling importance score , EC shows very good power to identify interactions . We have tested other classifiers , such as decision trees , and have found little sensitivity to the choice of classifier . However , the method for optimizing the importance score coupling will be a focus of future research . The genome-wide testing of DNA sequence variants for association with complex diseases opens up the possibility of identifying gene–gene interactions and even networks of interacting susceptibility loci . However , this network or pathway level view of SNPs affecting the expression of a phenotype will only be meaningful if analytical methods can identify gene–gene interactions . The EC filter is conducive to a pathway-level analysis because it accounts for the context of all SNPs when computing the relevance of a specific SNP to the phenotype . Furthermore , when coupled with network analysis such as GAIN , the collection of SNPs enriched for interactions may be modeled on a global/pathway level . We demonstrated the ability of EC network analysis to identify interactions between SNPs , the most common form of genetic variant , but EC is also applicable to gene expression data and the emerging CNP . By treating attributes as real-valued variables , gene expression data can be analyzed for interactive associations with a phenotype , and CNPs could be treated as discrete or real-valued to avoid converting raw intensities to genotypes . EC can be used for attribute selection in other domains of bioinformatics where statistical interactions may be significant , such as identifying biophysical properties of amino acids that predict protein binding .
To compare the performance of each analytical method , replicate data sets for the genetic models in Figure 1 were created with the genomeSIM software package [32] . The genomeSIM software was developed as a realistic , forward-time population simulation algorithm that allows the user to specify many evolutionary parameters and to control evolutionary processes . In the simulation , an initial population of diploid individuals is randomly created and individuals cross by contributing one chromosome to the offspring . These crosses form the next generation and the process repeats until the specified number of generations has occurred . In the final generation , summing across the binary chromosome pairs at each position produces genotypes for the individual . Disease status is assigned by the probability of disease for each genotype or genotype combination as defined in the penetrance function . Because LR is able to fit additive and other low order effects as well as interactions , we compare the filter methods in this paper with an LR with L2-regularization to fit gene interaction models [31] . As the number of markers in a genetic association data set grows it becomes increasingly unlikely that an exhaustive set of tests would be feasible , so a step-wise approach seems to be a reasonable approach for comparison with other filter methods . The authors in Ref . [31] implemented this method as an R package called stepPLR , which uses forward selection followed by backwards deletion for variable selection . In each forward step , a factor or interaction of factors is added to the model . In the backward step , factors are deleted beginning with the largest model from the forward steps . In our application , we selected the regularization parameter by cross-validation then built models based on the Bayesian information criterion . Relief-F is an extension of Relief , a heuristic machine learning method for estimating the quality of variants according to their ability to separate samples into classes . The following details of the algorithm apply to both Relief and Relief-F , then below we point out the differences . Consider a set of genetic variants ( e . g . , SNPs ) G , where each genetic variant gi in this set can be in one of the genotype states from the set {0 , 1 , 2} , corresponding to the homozygous for the common allele , heterozygous , and homozygous for the minor allele . In Relief , the weight of each attribute gi is initially set to zero ( W[gi] = 0 ) and for randomly selected samples ( or for all samples if desired ) the nearest hit and miss are computed with the chosen distance function ( metric ) and W[gi] is recursively updated according to how well the attribute can separate near hits and misses . Given a sample from one class , the nearest hit is defined as the nearest sample or individual from the same class as the sample of interest , where nearness in the SNP space is defined below . The nearest miss is the nearest sample from the opposite class . The selection of the nearest hit/miss is crucial to the success of Relief-F to find strong attribute dependencies because nearness is defined in the space of all SNPs as opposed to a single SNP at a time . For a given sample S ( or individual ) with nearest hit H and nearest miss M , the following equation is used to update the weight of each SNP gi: ( 2 ) This is repeated for m samples selected randomly or exhaustively . Division by m in Eq . ( 2 ) ensures that the weight of each attribute lies between −1 and 1 . For SNP gi , the difference function between samples Sj and Sk is ( 3 ) where genotype ( g , S ) means the genotype of SNP g for sample S . Eq . ( 3 ) is used also for calculating the distance between samples to find the nearest neighbors . The total distance is the Manhattan distance , or the sum of distances over all SNPs . The importance score W of a genetic variant gi is recursively updated for each individual , or sample S , in the population . Equation ( 2 ) rewards attributes that yield a large separation between the given sample and the nearest sample from the other class ( misses , M ) and penalizes attributes that give large separations between the given sample and the nearest sample from the same class ( hits , H ) . For example , if the separation of a sample from its nearest hit is the same as its separation from its nearest miss then the contribution to the weight of the attribute is zero because it does not contribute to the classification of the sample . In our algorithm , we use Relief-F , an extension of Relief that enables it to handle noisy and incomplete data sets and to deal with multi-class problems . The main difference from Relief is that Relief-F searches for the K nearest hits and misses instead of the single nearest hit and miss , which gives greater robustness with respect to noise . We used K = 10 nearest neighbors and exhaustive selection of samples . For more details on Relief-F , see [37] . We use the Relief-F feature-weighting algorithm in our EC objective function ( discussed below ) because of its demonstrated ability to handle attribute interactions in genetic data [24] , [26] . The iterative removal of the worst attributes in the evaporative formalism is the key to countering the devaluation of Relief-F importance scores of relevant SNPs due to the context of noise variants . As a control in the Results section , we compare EC with an iterative Relief-F called tuned Relief-F ( TuRF ) [24] . In our original construction of EC , we used Information Gain ( IG ) as the main-effect contribution ( the entropy term S ) to the information free energy score [24] , [26] . This was a natural choice for the evaporation formalism because of the basis of IG in information entropy . Although we show that Random Forest ( RF ) is not particularly good at identifying purely epistatic interactions ( see Results ) , it performs very well when identifying main effect variants that elude many standard methods ( e . g . , IG , chi-square , LR ) . Thus , we integrate the Random Forest importance ranking as the main-effect component ( S ) to the EC score ( discussed below ) . We use a version of RF known as PARF ( parallel RF ) that has been parallelized in Fortran 90 [36] . RF is a collection of decision tree classifiers , where each tree in the forest has been trained using a bootstrap sample of individuals from the data and each split attribute in the tree is chosen from among a random subset of attributes . Classification of individuals is based upon aggregate voting over all trees in the forest . Each tree in the forest is constructed as follows from data having N individuals and M explanatory attributes: Repetition of this algorithm yields a forest of trees , each of which has been trained on bootstrap samples of individuals . Thus , for a given tree , certain individuals were left out during training ( on average for a large number of samples , the fraction 1-1/e ) . Prediction error and attribute importance was estimated from these “out-of-bag” individuals . In RF the out-of-bag ( unseen ) individuals are used to estimate the importance of particular attributes according to the following logic: If randomly permuting values of a particular attribute does not affect the predictive ability of trees on out-of-bag samples , that attribute is assigned a low importance score . If , however , randomly permuting the values of a particular attribute drastically impairs the ability of trees to correctly predict the class of out-of-bag samples , then the importance score of that attribute is high . Tree methods are suited to dealing with certain types of genetic heterogeneity because splits near the root node define separate population subsets in the data . RF capitalizes on the established benefits of decision trees and has demonstrated excellent predictive performance when the forest is diverse ( i . e . , trees are not highly correlated with each other ) and composed of individually strong classifier trees [20] , [21] . By running out-of-bag samples down entire trees during the permutation procedure , weak attribute interactions are taken into account when calculating importance scores , since class was assigned in the context of other attribute nodes in the tree . However , RF has limited ability to identify strong interaction ( pure epistatic ) effects ( see Results section ) . An approach for improving the ability of RF to identify interactions can be found in Ref . [25] . The author found a slight increase in the performance of RF when several attribute evaluation measures , including Relief-F , were used as the split selectors for building the trees instead of only the Gini index . Ref . [25] used classification accuracy as the performance measure , but in the current paper we are more interested in the power to identify specific genetic variants that predict the phenotype variable . Rather than integrate Relief-F into RF as the split selector , the EC approach used in the current study computes the RF importance score ( with the Gini index ) and computes the Relief-F score outside of RF then couples them into a composite importance score . We introduced Evaporative Cooling ( EC ) as a machine learning method for feature selection in Ref . [26] . As illustrated in Figure 5 , the heuristic used in our new EC machine-learning algorithm is the evaporation of a collection of atoms to reach equilibrium by balancing low energy ( E ) and high entropy ( S ) via the temperature ( T ) to minimize the free energy , F = E-TS . The physical process of evaporative cooling was first proposed as an experimental technique for cooling a small gas of atoms by [38] . The experimental method consists of the selective removal of atoms in the high-energy tail of the thermal distribution and the collisional equilibration of the remaining atoms . The combination of atom selection and collisions increases the phase-space density and can greatly reduce the temperature of a gas . In the EC machine learning analogy , each atom represents a variant with genotype states whereby each variant contributes quantities analogous to energy and entropy to the relevance to the phenotype . The orange highlighted SNP in Step 0 of Figure 5 has genotype states corresponding to homozygous for the C allele ( CC ) , homozygous for the T allele ( TT ) and heterozygous ( CT ) . Each SNP makes a contribution to the “information free energy , ” F = E-TS , which quantifies the relevance of a collection of SNPs to the phenotypic variable . It is the goal of EC to minimize this quantity . The contributions to F of SNPs that are less relevant to the phenotype are positioned higher in the trap ( parabola ) , and these SNPs are allowed to escape the trap as the top of the trap is lowered . The key mechanism of EC is the balance of statistical interactions ( E ) and independent effects ( S ) via the “information temperature” T as noisy variants ( SNPs unrelated to the phenotype ) are evaporated ( iteratively filtered ) from the full collection of SNPs in the trap , leaving behind a subset of SNPs enriched for relevance to the phenotype . An important advantage of the EC formalism is the ability to assimilate alternative SNP relevance scores through the coupling constant T . In the present study we couple RF and Relief-F to boost the performance over each attribute importance score alone . Figure 5 gives an overview of the EC feature selection algorithm . At top left ( Step 0 ) , all N SNPs in the data set are represented as a gas of atoms in a fictitious trap , where more energetic SNPs ( red , top ) are poorly associated with the phenotype and “colder” SNPs ( blue , bottom ) are more closely associated with the phenotype . Relevance is determined by the attribute importance score F = E-TS , where E is the Relief-F score and S is the RF score , both transformed to be on the same scale with a range between 0 and 1 . Relief-F is further transformed so that a SNP with a low Relief-F score is more important . The information temperature is initialized to T = 1 ( least biased assumption ) so that the main effect and interaction terms of the attribute quality score are equally coupled . In Step 1 , an ensemble of gases is created from the initial set of SNPs by variation D of the information temperature T around the initial value [we use the range D = 22] . Since each collection in the ensemble uses a different coupling T+D , the rank order of the SNPs will differ in general . Thus , each collection of SNPs in the ensemble will have a different set of worst SNPs removed , indicated by an X in Step 1 , corresponding to different perturbations D of the information temperature T . In Step 2 the new value of the temperature and the particular SNPs removed are determined by the collection of remaining SNPs generated in Step 1 that yield the highest classification accuracy . We use a naïve Bayes classifier as we have found little sensitivity to the type of classifier . The goal of Step 1 is to locally search for the information temperature that removes the worst attributes , and in Step 2 the worst SNPs are evaporated , or permanently removed . In Step 3 the stopping criterion is checked . If the target number of SNPs ( Ntarget ) has not been reached ( “If No” ) , the evaporation procedure is repeated . In the example shown in Figure 5 , the new temperature would become T = T+D2 from Step 2 . Then iteration would continue with Step 1 with the SNPs ranked according to the new attribute importance score calculated by perturbing about this new temperature . The recalculation of F after the removal of noise attributes at each evaporation step is primarily motivated by the context dependence of Relief-F , which can lead to sensitivity to noise variants . However , we also find that RF benefits from the recalculation of importance scores . If on the other hand in Step 3 the number of SNPs is equal to or less than the target number of SNPs specified by the user ( “If Yes” ) , then the stopping criterion has been met and the remaining SNPs become the final collection of “cooled” SNPs that are most relevant to the phenotype . This final collection of SNPs is depicted as a frozen network of interacting attributes , which is inferred as a genetic-association interaction network ( GAIN ) of the phenotype ( discussed below ) . Just as evaporative cooling of an atomic gas increases the phase space density of the gas by repeatedly removing the most energetic atoms , the goal of EC feature selection is to alleviate the curse of dimensionality [39] by increasing the feature space density through the iterative removal of the genetic variants with the most noise . Relief-F makes the detection of interactions of order m computationally efficient because the complexity with respect to the number of SNPs , n , is O ( n ) , versus O ( nm ) for an exhaustive search . The final number of SNPs , Ntarget , is chosen based on a permutation strategy discussed below . The SNPs selected by EC are enriched for interactions as well as main effects , but EC does not predict which specific SNPs may be interacting . In order to characterize specific interactions among the top EC-ranked SNPs , we infer a genetic association interaction network ( GAIN ) . GAIN is based on II [28] between three attributes ( in this case , between two regular attributes A and B and the class attribute C ) : ( 4 ) where I ( A;C ) and I ( B;C ) are the information gained about the phenotype ( C ) when locus A and locus B , respectively , are measured . The quantity AB is a joint attribute constructed from attributes A and B with states given by the Cartesian product of the states of A and B . II is then the gain in class information obtained by considering A and B jointly beyond the class information that would be gained by considering variables A and B independently . We use the II ( Eq . 4 ) as the connection strength of each edge in the GAIN ( Figure 4 ) . Thus , each edge represents the increase in information about the phenotype achieved by considering the two SNPs jointly compared to the expected increase in information with the assumption of independence between the SNPs . We emphasize that a connection between SNPs in a GAIN is specific to the given phenotype because it measures the correlation between two SNPs that influences association with the phenotype . We have made the Java software freely available for generating the GAIN results . We built network visualization into the software tool , but to create Figure 4 we used the export option in the GAIN software for subsequent visualization in Cytoscape [40] , a freely distributed software tool for network visualization and annotation . A challenge for non-parametric methods like EC is assessing the statistical significance of a relevant SNP or , in the case of GAIN , a significant interaction between SNPs . We use a random permutation approach to determine a statistically significant threshold or cutoff for selecting the top EC SNPs and the top interaction pairs for GAIN . For EC we generate a distribution of irrelevant SNPs by randomly selecting SNPs with replacement and then calculating their EC score after randomly permuting the genotypes of the chosen SNP . From the resulting distribution of irrelevant SNPs , we determine the EC threshold by selecting the EC score such that only a fraction a of the irrelevant scores are more extreme . To select the interaction strength threshold for displaying GAIN edges , we calculate the II for randomly permuted pairs of SNPs . From the resulting non-interaction distribution of II scores , we use the same process to choose the threshold as we used for selecting the EC score threshold . | Susceptibility to many diseases and disorders is caused by breakdown at multiple points in the genetic network . Each of these points of breakdown by itself may have a very modest effect on disease risk but the points may have a much stronger effect through statistical interactions with each other . Genome-wide association studies provide the opportunity to identify alleles at multiple loci that interact to influence phenotypic variation in common diseases and disorders . However , if each SNP is tested for association as though it were independent of the rest of the genome , then the full advantage of the variation from markers across the genome will be unfulfilled . In this study , we illustrate the utility of a new approach to high-dimensional genetic association analysis that treats the collection of SNPs as interacting on a system level . This approach uses a machine-learning filter followed by an information theoretic and graph theoretic approach to infer a phenotype-specific network of interacting SNPs . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"genetics",
"and",
"genomics/genetics",
"of",
"the",
"immune",
"system",
"genetics",
"and",
"genomics/complex",
"traits",
"genetics",
"and",
"genomics/disease",
"models",
"genetics",
"and",
"genomics/genetics",
"of",
"disease",
"genetics",
"and",
"genomics/population",
"genetics"
] | 2009 | Capturing the Spectrum of Interaction Effects in Genetic Association Studies by Simulated Evaporative Cooling Network Analysis |
Vibrio cholerae has evolved to adeptly transition between the human small intestine and aquatic environments , leading to water-borne spread and transmission of the lethal diarrheal disease cholera . Using a host model that mimics the pathology of human cholera , we applied high density transposon mutagenesis combined with massively parallel sequencing ( Tn-seq ) to determine the fitness contribution of >90% of all non-essential genes of V . cholerae both during host infection and dissemination . Targeted mutagenesis and validation of 35 genes confirmed our results for the selective conditions with a total false positive rate of 4% . We identified 165 genes never before implicated for roles in dissemination that reside within pathways controlling many metabolic , catabolic and protective processes , from which a central role for glycogen metabolism was revealed . We additionally identified 76 new pathogenicity factors and 414 putatively essential genes for V . cholerae growth . Our results provide a comprehensive framework for understanding the biology of V . cholerae as it colonizes the small intestine , elicits profuse secretory diarrhea , and disseminates into the aquatic environment .
The human intestinal pathogen Vibrio cholerae is a natural inhabitant of the aquatic ecosystem . V . cholerae is a halophile that normally resides in coastal waters and brackish estuaries , however pathogenic strains can persist in bodies of fresh water for extended periods of time in cholera endemic areas [1] . Human consumption of V . cholerae in contaminated food and water leads to colonization of the small intestine , extensive bacterial multiplication , and cholera . Activation of the virulence gene cascade upon colonization leads to the production and secretion of the cholera toxin ( CT ) into the intestinal lumen ( reviewed in [2] ) . CT induces the secretion of water , electrolytes and mucin into the bowel resulting in profuse secretory diarrhea , known as rice water stool ( RWS ) [3] . CT-induced diarrhea can cause death by dehydration within hours in the absence of aggressive rehydration therapy [4] . Diarrheal expulsion of V . cholerae from the host into the aquatic environment leads to dissemination and transmission of cholera , thus completing the pathogen life cycle . Host-passaged V . cholerae are an order of magnitude more infectious than in vitro grown bacteria , and this hyperinfectious state is maintained for a limited time in fresh water [5] , [6] . Contamination of household food and/or water with hyperinfectious V . cholerae can lead to rapid spread of the disease and fuels epidemics [7] , [8] . When RWS V . cholerae are shed into fresh water , the bacteria encounter a drastic drop in osmolarity , temperature and nutrient availability [6] . Transcriptional analysis of RWS V . cholerae transitioned into pond water revealed repression of protein synthesis and energy metabolism genes and induction of phosphate and nitrogen scavenging genes , consistent with low levels of carbon , phosphate and nitrogen in the pond [6] . However , the importance of these genes for dissemination was not examined further . Indeed , very little is known about the genetic factors important for V . cholerae survival in fresh water after host passage . Glycogen storage is one of the few processes described that enhance dissemination of V . cholerae [9] . Glycogen is a branched glucose polymer that accumulates in stationary phase and is used as a carbon and energy source under nutrient limiting conditions , such as the aquatic environment . Glycogen storage granules have been detected in V . cholerae isolated from RWS , and some of the glycogen storage and utilization genes have been determined to be important in dissemination and transmission [9] . A few studies have suggested that host-passaged V . cholerae exit the host in a conditioned state to better survive the transition into fresh water [10] , [11] . The data suggest that these host-passaged V . cholerae are similar to bacteria in stationary phase [11] , [12] . Until recently , a suitable animal model to study dissemination into the aquatic environment was lacking . Although V . cholerae colonize the small intestine of the infant mouse , CT-induced secretory diarrhea is not produced . An infant rabbit model of infection was recently improved to give consistent and reproducible symptoms of secretory diarrhea that more closely resemble human disease [13] . In this study , we utilized the infant rabbit model of cholera to study both pathogenesis and subsequent dissemination into the aquatic environment . We sought to determine on a genome-wide basis the genetic determinants necessary for survival in both conditions . Using the Tn-seq method [14] , [15] , which combines saturating transposon mutagenesis with massively parallel sequencing ( MPS ) of transposon junctions , we quantitatively measured the fitness of each gene during both infection of the host small intestine and dissemination to a fresh water environment .
Eleven mini-Tn5 ( mTn5 ) transposon libraries were constructed in V . cholerae O1 serogroup , El Tor biotype strain E7946 , representing over 100 , 000 unique transposon insertions as determined by MPS . From the Tn-seq data , we identified 414 genes that lacked Tn insertions , and therefore are deemed putatively essential for V . cholerae viability in rich medium ( Figure 1 and Table S1 ) . With 100 , 000 unique transposon insertions in the remaining 3471 non-essential genes ( average of one insertion every 36 bp ) , our total library coverage is likely to be near saturating . Comparison of our essential gene list to those previously reported in V . cholerae C6706 [16] , [17] , showed 87% ( 361/414 ) overlap ( Table S1 ) . The libraries were tested in three selective conditions including rich medium ( in vitro ) , host ( in vivo ) , and dissemination from host to aquatic environment . The orogastric infant rabbit model of V . cholerae infection was used to represent the host selective condition [13] , fresh pond water was used to represent the aquatic environment [6] , [10] and Luria Bertani ( LB ) broth was used for the rich medium . After the transposon library was subjected to a selective condition ( Figure 2 ) , the surviving population was outgrown a limited number of generations in LB broth to achieve high cell density . This outgrowth eliminated the potential for DNA contamination from dead V . cholerae cells . Genomic DNA ( gDNA ) was isolated from each outgrown sample and processed as illustrated in Figure 2 ( also see Experimental Procedures ) . The final PCR products contained sequences corresponding to the transposon , gDNA flanking the transposon , Illumina-specific sequences on each end necessary for MPS , and a unique 6 bp barcode embedded in one end to identify the sample . Samples were multiplexed and subjected to MPS to determine the frequency of each transposon insertion mutant in both the input and output populations . The fitness contribution of each non-essential gene in the genome ( hereafter referred to as “a gene's fitness” ) was calculated for each selective condition as described [15] , where each gene's fitness represents a measurement of the net growth ( in LB broth or the host ) or death ( in pond ) of that gene disruption relative to the bulk population . In order to be able to equate fitness with growth rate in the rich media and host conditions , we needed to determine the population expansion during each experiment . This is easily determined in LB broth by enumerating the total increase in colony-forming units ( CFU ) , however bacterial growth is more difficult to assess during infection due to the stochastic and host-induced loss of bacteria during the course of infection . In order to determine the expansion of V . cholerae during host infection we utilized a low copy , temperature-sensitive plasmid that is replication defective at physiological temperatures [18] . As the bacteria divide and replicate in vivo , the temperature sensitive plasmid is lost by segregation [19] . After the course of host infection , we compare the plasmid positive ( KanR ) and plasmid negative ( KanS ) populations to determine the ratio of plasmid loss . By fitting the ratio of plasmid loss in vivo to an in vitro standard curve at physiological temperature , the expansion of the in vivo population was estimated and used in our fitness calculation . Stochastic loss of bacteria due to narrow bottlenecks in the selective conditions can lead to noisy fitness calculations with large standard deviations . To widen the bottleneck in the in vivo selection , specifically during passage through the stomach , the following steps were taken; 1 ) we pre-conditioned V . cholerae by acid-tolerizing them for 1 hr prior to intragastric inoculation [20] , 2 ) the inoculum was suspended in an alkaline bicarbonate-buffer , and 3 ) animals were pre-treated with Cimetidine to block stomach acid production as described [13] . The bottleneck , defined as the percentage of neutral genes lost stochastically , was never greater than 15% and the median ranged from 0–7% depending on the selective condition ( Table 1 ) . To reduce the number of false positives and false negatives , the final fitness values of genes obtained under each selective condition were calculated from multiple insertion sites per gene ( ≥5 ) from 2–4 independent experiments representing a total of 9–21 biological replicates ( Table 1 ) . These stringent criteria resulted in exclusion of 6% of genes in LB that had less than 5 insertion sites but were not determined to be essential ( 11% ) ( Figure 1 ) . Of the remaining 83% of genes , we were able to calculate fitness values in one or more selection conditions ( Table 1 and Table S2 ) . A fitness value of 1 . 0 is neutral , less than 1 . 0 is disadvantageous , and greater than 1 . 0 is advantageous . To determine if a fitness value was statistically different from 1 . 0 , a one-sample student's t test with Bonferroni correction for multiple testing was used . To determine if a fitness value is specific for the selective condition , we required the fitness value to be statistically different than the fitness value obtained in LB and required at least a 0 . 15 fitness value difference between the two fitness values . In addition , any gene with a fitness value less than 0 . 82 in LB was considered to have a general growth defect and eliminated from the host and pond gene lists . These stringent requirements yielded between 80–148 genes required for each condition ( Table 1 and Table S3 ) . The percentage of genes within each class is shown in Figure 1 . In contrast to the more commonly used infant mouse model , the orogastric model of infection in the infant rabbit better approximates severe human cholera because of the large volume of secretory diarrhea that results [13] , [21] . Prior to diarrheal excretion , the luminal fluid and V . cholerae shed from the small intestine accumulate in the cecum of the infant rabbit [13] . Collection of the cecal fluid provided essentially pure cultures of host-passaged V . cholerae that was used to study both pathogenesis and dissemination into the aquatic environment . The fitness values calculated for genes obtained from the cecal fluid correlated well with those calculated using the small intestine ( Figure 3 ) , indicating that the former population is highly representative of the latter . Therefore , all genes identified as important for survival in the host were collectively determined by combining the statistically significant gene lists from both the cecal fluid and small intestine ( Table 2 and Figure 1 ) . There were 133 genes identified as being important for survival specifically in the host environment ( Table 2 and Table S3 ) . Over half of these genes ( 76 genes; 57% ) were new factors that had not been previously implicated in V . cholerae infection , and thus the data set provides a wealth of information for future studies in V . cholerae pathogenesis . The remaining 57 genes ( 43% ) had previously been shown to be important for V . cholerae colonization in the infant mouse model of infection , including the well-known virulence factors: ToxR , ToxS , ToxT , the TCP biosynthetic operon , and the lipopolysaccharide O-antigen biosynthetic operon ( Table S3 ) . Classification of the known or hypothesized functions of the genes important for survival reveals that a large number are required for purine and pyrimidine biosynthesis , O-antigen biosynthesis , and amino acid metabolism ( Table 2 and Table S3 ) . The remaining genes are spread across many functional classes including phosphate acquisition , post-translational modification , fatty acid metabolism , and transporters . Since the majority of the population has a neutral phenotype , some classes of mutations would be missed due to the ability of the bulk population to complement the mutant defect in trans . For example , the cholera toxin ( CT ) genes ctxA and ctxB had neutral fitness values since the remaining population secreting CT into the small intestine can complement their deletion in trans . In addition to calculating genome-wide fitness values for host infection , we also determined fitness survival values for dissemination into the aquatic environment . We isolated host-passaged V . cholerae from the infant rabbit cecal fluid , which is similar in composition to human RWS [13] . From each infected animal , we obtained between 0 . 2–1 . 2 ml of cecal fluid containing ∼108 CFU/ml of V . cholerae . These bacteria were pelleted , resuspended , and diluted into pond water and then incubated at 30°C for 2 days statically in an open container . Concurrently , transposon libraries grown in LB broth overnight to stationary phase were also pelleted , resuspended , and diluted into pond water . Direct comparison between host-passaged and LB stationary-phase bacteria using competition experiments on the bulk populations showed no significant difference in pond survival at 2 days ( Figure 4A ) , however host-passaged V . cholerae show a significant survival advantage at 8 days ( Mean competitive index [CI] = 18 ) . In contrast to host-passaged and stationary phase bacteria , exponential growth-phase V . cholerae from LB broth do not survive well in pond water at 2 days ( Figure 4 ) . After 3 days in pond water , the survival of stationary phase bacteria also drops severely in contrast to host-passaged bacteria ( Figure 4B ) . To avoid severe bottleneck restraints incurred at longer time points , we examined genome-wide fitness after 2 days post-dissemination of both host-passaged and stationary phase bacteria . Statistical comparison to both the neutral fitness value of 1 . 0 and to the fitness value obtained in LB revealed 165 genes that are important for survival in the pond environment ( Table 2 and Table S3 ) . These included both host-passaged and stationary phase into pond selections ( Table 2 ) . Functional classification indicates that a large number of these genes have known or hypothetical roles in energy production and conservation , cell wall and outer membrane biogenesis , electron transport , flagellar biosynthesis , transcriptional regulation , transporters and hypothetical proteins . Since this is the first genome-wide screen to determine factors important for dissemination and survival in the aquatic environment , almost all of the genes identified are new . In order to validate the results of the Tn-seq screen , a mini-library of 35 gene deletions representing host-specific , pond-specific , neutral , and dual specificity factors covering a wide range of fitness values was constructed and tested collectively in the selective conditions ( Table S4 ) . This strategy allowed us to test a large number of genes while greatly reducing the number of animals needed for the experiment . These 35 deletions were constructed by combining allelic exchange by natural transformation with Frt/Flp recombination [22] , resulting in an identical 81 bp in-frame “scar” between the start and stop codons in each deletion ( Figure 5A ) . This 81 bp sequence flanks unique sequence at each deletion locus , with the latter serving as a barcode that can be used to track each mutant in the mini-library by MPS using an essentially identical protocol to the original Tn-seq screen ( Figure 5A , Figure 2 ) . In addition to re-testing these 35 deletion mutants in infant rabbits , pond , and rich media , we also infected infant mice with the mini-library to compare the two host models . For the infant mouse experiment , a CI was calculated based on MPS reads normalized to a neutral gene deletion in both the input and output pools . Further validation of the genes important for pond survival also included 1∶1 competitions in pond water with a wild-type strain ( Figure 5C and Table S4 ) . Direct comparison between the fitness values obtained for all conditions in the Tn-seq screen to the fitness values obtained with the deletions revealed a strong correlation ( Figure 5B ) ; 97% ( 34/35 ) of the deletions were confirmed for the host selection , and 91% ( 32/35 ) of the deletions were confirmed for the pond selection ( Table S4 ) . The percentage of false positives are in line with 5% false discovery rate that was imposed in the Tn-seq statistical calculations [14] . Nine out of 11 infant rabbit host-specific factors were also shown to be important for virulence in the infant mouse ( Table S4 ) , suggesting that these two host environments are highly similar . The two factors that appear to be rabbit-specific are gltA and mtlD , encoding a citrate synthase and a mannitol metabolic protein , respectively . Interestingly , gltA is also important for survival in the aquatic environment ( Figure 5C and Table S4 ) . The necessity of these genes in the infant rabbit model but not the infant mouse model , suggests there are differences in carbon and energy sources available to V . cholerae in the two hosts . The VarS sensor kinase and cognate VarA response regulator make up a two component system that regulates the transcription of virulence genes in V . cholerae [23] , [24] , [25] and controls both the expression and the activity of the HapR quorum sensing regulator [26] , [27] . Both varA and varS deletion strains have colonization defects in mice [23] , [25] . In our original screen , transposon insertions in both varS and varA resulted in fitness defects in both the host and in the pond , but not during growth in rich media ( Figure 6A ) . To validate the pond phenotype we constructed a varS deletion strain and competed it 1∶1 with a wild-type strain in pond water for 48 hrs ( Figure 6B ) . A ΔvarS suppressor strain ( ΔvarS** ) which phenocopies the wild-type strain was isolated during the construction of ΔvarS , due to nutrient-limited conditions required for natural transformation . Whole genome resequencing of ΔvarS** revealed a single nucleotide change located in the start codon ( ATG→ATA ) of csrA ( carbon storage regulator gene A ) . However , there is a putative alternative start codon ( TTG ) 6 bp downstream with an appropriately spaced Shine-Dalgarno sequence , which we speculate provides reduced expression of CsrA . CsrA is a global post-transcriptional regulator that has been shown to control many processes in E . coli including glycogen synthesis and storage ( Reviewed in [28] ) . The VarS/VarA homologs BarA/UvrY in E . coli antagonize CsrA by activating transcription of an inhibitory sRNA , csrB . The binding and sequestration of CsrA by csrB promotes translation of downstream targets , including the mRNA for the glycogen biosynthetic genes glgCAP [29] . Thus , the varS** suppressor is predicted to restore repression of CsrA thus allowing for expression of downstream CsrA-repressed genes . Since glycogen storage is known to be important for V . cholerae survival in the aquatic environment [9] , we measured and compared the amount of glycogen in a wild-type strain of V . cholerae to both ΔvarS and ΔvarS** ( Figure 6C ) . The ΔvarS strain has significantly less glycogen stored than both the wild-type strain and ΔvarS** , which likely accounts for the survival defect in the pond ( Figure 6B ) .
The water-borne pathogen V . cholerae encounters disparate environmental conditions when it transitions between the human small intestine and the aquatic environment . Understanding the genetic determinants important for survival during each stage of the life cycle of this organism is key to understanding the dynamics of cholera outbreaks and persistence during and in between outbreaks . In this study , we determined genome-wide fitness in a mixed population for gene disruptions during both host infection and dissemination into a pond water aquatic environment . By using a host model that closely resembles human cholera , we have obtained new information about V . cholerae pathogenesis and for the first time have been able to identify factors important for dissemination from watery stools into the aquatic environment . Overall our study is unique in several respects: first , it presents the first genome-wide data set from a saturating mutant screen that identifies the roles of genes in a V . cholerae infection; second , it is the first genome-wide data set on genes important for dissemination from the host into an aquatic environment; and third , it represents the first genome-wide screen done using the infant rabbit model . The current knowledge of V . cholerae virulence resides mostly in data obtained from a handful of non-saturating mutagenesis screens done using the infant mouse model of infection [20] , [30] , [31] , [32] . While V . cholerae is able to colonize the small intestine of infant mice , it does not elicit the same fatal secretory diarrheal response as seen in human cholera; while however , this pathology is replicated in the infant rabbit host model [13] . A recent RNA-seq study comparing the V . cholerae transcriptome during infection of the infant mouse or infant rabbit found only partial overlap between the two hosts [33] . Despite the differences in V . cholerae gene expression and disease outcome between the two host models , we identify 57 virulence factors in the infant rabbit model that had previously been shown to be important in the infant mouse model . Among these previously identified factors are the genes required for TCP biogenesis , O-antigen synthesis , phosphate acquisition , quorum sensing regulators , and key virulence transcriptional regulators including ToxR , ToxS , ToxT , VarS , VarA and AphA . As part of the validation of our Tn-seq screen we compared the roles of 11 genes in colonization of the infant rabbit to that of the infant mouse . A majority of the genes ( 9/11 ) , were important for colonization in both models . Thus , our data indicate that the requirements for infection of these two seemingly dissimilar animal hosts are remarkably comparable . Among the notable differences in gene requirement in the two animal models are the well-known virulence factors TcpP , TcpH , and the accessory colonization factors AcfA-D , which gave mild or neutral phenotypes in the infant rabbit host . In the infant mouse model and in laboratory conditions that stimulate virulence gene expression , TcpPH is known to function with ToxRS to co-stimulate transcription of toxT , which encodes the main transcriptional activator of the ToxR virulence regulon [34] , [35] . Interestingly , a recent study using the infant rabbit model showed that transcription of tcpPH is not induced during infection when compared to in vitro transcription [33] . The same study also revealed that the biosynthetic genes for vibriobactin , an iron siderophore important for infant mouse colonization , are not induced in the infant rabbit , and that instead heme may be used as the main iron source during infant rabbit colonization [33] . We confirm this hypothesis at the phenotypic level , by showing that vibriobactin genes are dispensable during infant rabbit colonization , while a heme transporter ( VC2056 ) is needed . It has been previously demonstrated that V . cholerae incorporates exogenous fatty acids from host bile to alter the membrane phospholipid profile [36] . Long chain fatty acid uptake in E . coli requires three known proteins , FadL , FadD and PlsB . FadL transports exogenous long chain fatty acids from the environment into the cell where the acyl-CoA synthetase FadD converts the fatty acids into acyl-CoA thioesters and PlsB , an acyl transferase processes them into phosphatidic acid , the precursor for phospholipid synthesis [37] , [38] , [39] . We identify one of the three homologs of FadL ( VC1043 ) in V . cholerae as specifically required for dissemination into the aquatic environment from the host . The other two FadL homologs had neutral phenotypes for all conditions . FadD ( VC1985 ) and PlsB ( VC0093 ) are required for both pathogenesis and dissemination in V . cholerae . The identification of FadD , FadL , and PlsB in our screen may highlight the importance of remodeling the lipid composition of the V . cholerae membrane during host infection prior to transition into the aquatic environment . This process may be part of a homeoviscous adaption required to survive the membrane stress encountered upon exposure to the host environment and to pond water . Examination of the fitness values obtained for the genes important for motility revealed a surprising result for infection . The overwhelming majority of flagellar biosynthesis genes and chemotaxis genes are dispensable for infection of the infant rabbit . There exist conflicting results in the infant mouse model for the role of motility and chemotaxis during host infection , while there are some reports that flagellar mutants are attenuated , there are other reports that indicate that flagellar motility is dispensable ( Reviewed in [40] ) . The extensive genome coverage of our data allows us to look at every single gene known to be required for flagellar biosynthesis and chemotaxis , and the results showed that almost all genes in the flagellar biosynthetic and chemotactic pathways show a neutral fitness value in both the small intestine and in the cecal fluid . Flagellar motility is thought to be important for establishing contact with the intestinal epithelium , which induces Cholera toxin mediated secretion of mucus , water , and electrolytes . It is possible that trans complementation by the bulk population ( i . e . , >99% of population is motile ) removes the need for flagellar motility during infection . In this scenario the flagellar motility mutants travel through the gastrointestinal tract without establishing contact with the epithelial cell layer , but may contact the mucus layer and benefit from the environment created by the bulk population . Alternatively , motility may be important under more natural , low dose inoculation scenarios , but this requirement might be obviated following the orogastric inoculation of a large bolus of V . cholerae , as was done in this study . In contrast to the in vivo environment , there is a requirement for some flagellar biosynthetic genes for full fitness in the aquatic environment . Although the flagellar biosynthetic genes are found in three distinct regions on the V . cholerae genome , the majority of the genes important for pond survival are all found at one chromosomal region ( VC2187–VC2208 ) and encode basal body , rod , ring , hook and filament genes , which are parts of the flagellum that have to be secreted through the flagella-specific type three secretion machinery [41] . It is unlikely that the lack of motility in these mutants is causing the survival defect since there are several additional non-motile flagellar mutants identified in our screen that do not have a survival defect in the pond , including important flagellar regulatory genes such as fliA , flrC and flrA . A more plausible explanation is that the survival defect of these flagellar biosynthetic mutants is instead caused by a loss of energy due to misregulation of the remaining flagellar components . The dissemination of V . cholerae from the host into the aquatic environment facilitates the transmission of cholera during epidemics and is also critical for the long-term survival of this pathogen in the environment . Previous studies in the adult rabbit ileal loop model show evidence of a stationary-phase dependent mucosal escape response , where V . cholerae detach from the host epithelial cell layer at the late stage of infection when bacterial cell density on the epithelium is high [11] . We therefore explored whether the physiological state of V . cholerae affected its ability to disseminate by comparing the survival of exponential-phase , stationary-phase , and host-passaged V . cholerae in pond water . The host-passaged and stationary phase bacteria had a significant survival advantage over exponential growth-phase cells . Intriguingly , host-passaged bacteria showed a long-term survival advantage over stationary phase cells . Using Tn-seq we were able to determine the fitness for 96% of V . cholerae non-essential genes during survival in the aquatic environment , resulting in the identification of 165 genes that play significant roles . Our data suggest that in order to survive the transition into the aquatic environment , V . cholerae must conserve energy and utilize its current cellular state rather than synthesize anew . For example , we found that several pathways involved in recycling and degradation of amino acids , proteins , and cell wall are important for V . cholerae fitness in the pond . V . cholerae will likely also need to scavenge nutrients and micronutrients that it can find in the environment , explaining the large number of transporters important for survival in the aquatic environment . One interesting family of transporters identified in our screen is the spermidine/putrescine ABC transporters . These polyamines can bind DNA , RNA and phospholipids and modulate cellular functions [42] . Disruption of potA , potC and potD , all of which are components of the spermidine/putrescine transporter , and speC , a putrescine biosynthesis gene all had fitness defects in the aquatic environment . We also identified and validated a set of genes , mdoG and mdoH , known to be involved in osmoprotection in proteobacteria [43] . These genes are required for periplasmic glucan biosynthesis , which are membrane-derived oligosaccharides that increase during low osmolarity and act as osmoprotectants . In conjunction with the transcriptional analysis of V . cholerae transitioned into pond water from rice water stool [6] , we identify a large number of energy production and conversion genes , the phosphate acquisition regulators PhoB and PhoR , along with GlnL and GlnE involved in nitrogen scavenging , to be important for survival in the aquatic environment . Our screen revealed that the two phosphate binding proteins PstS I and II are differentially required in the host and the aquatic environment , although the basis for this differential requirement is not known . Another regulatory system that we identify as important for fitness both in the host and in the aquatic environment but not in rich media is the VarS/VarA two component system . VarS/VarA is homologous to BarA/UvrY in E . coli , and has been previously identified in V . cholerae to be important for virulence and quorum sensing [23] , [26] . We now include a role for VarS/VarA in dissemination . In E . coli , BarA/UvrY have been shown to regulate glycogen storage through sRNA inhibition of the carbon storage regulator , CsrA [44] . Glycogen storage granules have been detected in V . cholerae isolated from RWS , and some of the glycogen storage and utilization genes have been determined to be important in dissemination [9] . We show that glycogen storage is decreased in a varS deletion strain , which can be suppressed by a mutation that decreases the expression of CsrA . These results indicate that VarS/VarA regulates glycogen storage through CsrA in V . cholerae , which affects survival in the aquatic environment . We also determined in our screen that the glycogen storage gene glgA and the glycogen degradation genes malQ and glgP are important for survival in both the host and the pond . In contrast , glgX , which is needed for breakdown of glycogen stores , is important specifically for pond survival . Glycogen catabolism helps V . cholerae meet its carbon and energy requirements during the stressful transition to nutrient-limited aquatic environments . A role for glycogen metabolism during infection of the comparatively nutrient rich intestinal environment is less obvious , but may be related to the integration of glycogen metabolism into central metabolism and bacterial physiology . These results highlight the important role that glycogen storage and catabolism play during the life cycle of V . cholerae . These data provide a comprehensive framework for understanding V . cholerae biology during the two critical stages of its life cycle; colonization of the small intestine , and dissemination into the aquatic environment . As such , these data comprise an important resource for future studies into the pathogenesis , dissemination and transmission properties of V . cholerae .
A streptomycin-resistant derivative of V . cholerae O1 serogroup , El Tor biotype strain E7946 [45] was used in this study . Single gene knockouts were constructed by utilizing the FRT/FLP recombinase method of deletion as described [22] . V . cholerae and E . coli were grown on LB agar or in LB broth at 37°C unless otherwise specified . Streptomycin ( Sm ) was used at 100 µg/ml , Carbenicillin ( Carb ) at 100 µg/ml , and Kanamycin ( Kan ) at 100 µg/ml . Strains utilized in this study are shown in Table S5 . Primer sequences for gene deletions are listed in Table S6 . Pond water was collected from Chandler pond ( latitute 42 . 345223 , longitude −71 . 164262 ) on 1/9/2011 and Ponkapoag Pond ( latitude 42 . 188406 , longitude −71 . 093241 ) on 4/7/2011 , 9/23/2011 , 6/4/2012 and 9/22/2012 . Previous analysis determined that the pond water samples had similar chemical composition as those obtained from a cholera endemic area in Southeast Asia [10] . The pUTmTn5Km2 plasmid in Sm10λpir was delivered to V . cholerae E7946 by mating with selection for mTn5Km2 transposition into the recipient genome using similar methods as previously described [20] . In summary , log phase E7946 and Sm10λpir+pUTmTn5Km2 were pelleted , resuspended to an OD600 = 8 . 0 , and mixed 1∶1 . Aliquots of 0 . 2 ml of the 1∶1 mating mixture were plated onto LB agar plates without antibiotic and incubated for 4 hrs at 37°C . After the 4 hr incubation , the LB plates were replica plated onto Sm100 Kn180 plates and incubated at 37°C for 12 hrs , and then replica plated again on Sm100 Kn180 plates for another 24 hrs at room temperature . Colonies were pooled ( 11 pools ranging from 5 , 000–80 , 000 CFU ) and resuspended in LB with 15% glycerol to an OD600 of 1 . 0 and frozen in 1 ml aliquots at −80°C . Library complexity was determined by Tn-seq . Libraries were prepared on two independent days . It is important to note that the mTn5Km2 transposon contains factor-independent transcriptional terminators on both transposon ends and therefore disrupts transcription of downstream genes in operons . Litters of 3-day old New Zealand White infant rabbits were obtained from a commercial source ( MillBrook Labs , Amherst , MA ) and housed together with their doe for the duration of the experiments . The dams and their litters were housed with food and water ad libitum and monitored in accordance with the rules of the Department of Laboratory Animal Medicine at Tufts Medical Center . The 3-day old infant rabbits were treated with 300 mg/kg of Cimetidine-HCL ( Morton Grove Pharmaceuticals ) by oral gavage 3 hrs prior to infection . Rabbits were orogastrically inoculated with ∼5×108 CFU of V . cholerae . To prepare the inocula , transposon library aliquots were thawed and diluted 1∶100 in LB containing Sm and Kan and grown at 37°C with aeration until late exponential phase ( OD600 = 0 . 5 ) . The bacteria were then pelleted and diluted 1∶2 in LB adjusted to pH 5 . 7 with HCl and grown for 1 hr at 37°C with aeration . At 1 hr , the acid-tolerized bacteria were pelleted and resuspended in 2 . 5% sodium bicarbonate buffer ( pH 9 ) to a final concentration of ∼109 CFU/ml . The infant rabbits are euthanized at 12 hr post-infection , when the infant rabbits exhibit signs of secretory diarrhea and Cholera disease . The cecal fluid was collected from the cecum by puncture and the contents were drained into a sterile petri dish . The small intestine was also collected and homogenized in sterile Phosphate Buffered Saline ( PBS ) . A low speed spin ( 500 RCF ) for 2 minutes removed cell debris . Both the cecal fluid and small intestine homogenate was serially diluted and plated on Sm , Km plates for bacterial enumeration . After collection , in vivo samples were minimally outgrown in LB by placing 0 . 1–0 . 3 ml of either cecal fluid or small intestine homogenate into 8 ml of LB and grown 2–6 hr at 37°C with aeration until turbid ( OD600 = 1 . 0 ) . Glycerol was added to a final volume of 20% and 2 ml aliquots were frozen at −80°C until gDNA could be prepped for Tn-seq . Both the culture prior to acid-tolerization and the final inocula used to infect the infant rabbits were diluted and outgrown in LB broth overnight at 30°C ( LB and LB-B , respectively ) . Samples were collected from both the starting population ( T1 ) and the end population ( T2 ) for both enumeration by CFU/ml and for Tn-seq . The LB samples were used to determine the putatively essential genes , the fitness of non-essential genes , and for comparison to gene fitness in pond . The LB-B fitness values were used for comparison to gene fitness in the host . The overnight stationary phase LB libraries were pelleted , washed twice with filter-sterilized pond water and then diluted 1∶100 in pond water and incubated at 30°C for 48 hrs . Host-passaged V . cholerae libraries in the cecal fluid were also placed in pond water for 48 hr at 30°C . First a low speed ( 500 RCF ) spin was used to remove large particulate matter and eukaryotic cells , the supernatant was collected and placed in a new microfuge tube and spun at 12 , 000 RCF to pellet the bacteria . The pellet was resuspended in 10 ml of pond water and incubated at 30°C for 48 hr . At 48 hr the bacteria were pelleted and resuspended in 8 ml of LB and grown to until OD600 = 1 . 0 , aliquoted with glycerol and frozen for Tn-seq . The transposon junctions were amplified from gDNA samples and subjected to MPS essentially as previously described [46] . Changes to these methods and primers are specified in the Supplemental Materials and Methods ( Text S1 ) . All read mapping and primary data analysis was done on the Tufts University Galaxy server ( http://genomics . med . tufts . edu/home/analysis ) using fitness calculation scripts nearly identical to those previously described [15] . Specific details of the fitness calculation for both the Tn-seq screen and the mini-library competition experiment are listed in the Supplemental Materials and Methods ( Text S1 ) . Overnight stationary phase cultures grown at 37°C with aeration of ΔlacZ and either wild type or a deletion strain were mixed 1∶1 , serially diluted , and plated on LB agar with 40 µg/ml 5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside ( X-gal ) . A volume of 0 . 3 ml of the 1∶1 mixture was pelleted by centrifugation and resuspended in 1 ml of filter sterilized pond water . A volume of 10 µl of the resuspended cell mixture was then put in 10 ml of filter sterilized pond water and incubated at 30°C standing for 48 hr . At 48 hr , the tube was vortexed rigorously , serially diluted and plated for CFU/ml on LB X-gal agar plates . The competitive index ( CI ) was calculated as the ratio of the mutant compared to the control strain normalized to the input ratio . A minimum of five biological replicates was used to calculate the average CI and a one-sample student t-test was used to determine significance . All animal experiments were done in accordance with NIH guidelines , the Animal Welfare Act and US federal law . Tufts University School of Medicine's Institutional Animal Care and Use Committee approved the experimental protocol “B2013-44” that was used for this study . All animals were housed in a centralized and AAALAC-accredited research animal facility that is fully staffed with trained husbandry , technical and veterinary personnel . | Cholera is a deadly diarrheal disease that spreads in explosive epidemics and is caused by the water-borne bacterium Vibrio cholerae . Pathogenic strains of V . cholerae can be found in both fresh and salt water estuaries in-between cholera outbreaks . Cholera infections are frequently derived from contaminated fresh water sources . In this study , we sought to determine on a genome-wide scale how V . cholerae is able to colonize and proliferate in the nutrient-rich environment of the small intestine , but then also survive dissemination and persist in the nutrient-limited aquatic environment . Using a host model that mimics the pathology of human cholera , we utilized genome-wide transposon mutagenesis and massively parallel sequencing of the insertion junctions to obtain the relative fitness of V . cholerae mutants during infection and dissemination . This extensive data set represents the first genetic screen of any kind to identify genes important for dissemination into the environment and has broad significance for understanding and controlling the spread and persistence of Vibrio cholerae and potentially other water-borne pathogens in the environment . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | Gene Fitness Landscapes of Vibrio cholerae at Important Stages of Its Life Cycle |
Deficiency of the protein Wolfram syndrome 1 ( WFS1 ) is associated with multiple neurological and psychiatric abnormalities similar to those observed in pathologies showing alterations in mitochondrial dynamics . The aim of this study was to examine the hypothesis that WFS1 deficiency affects neuronal function via mitochondrial abnormalities . We show that down-regulation of WFS1 in neurons leads to dramatic changes in mitochondrial dynamics ( inhibited mitochondrial fusion , altered mitochondrial trafficking , and augmented mitophagy ) , delaying neuronal development . WFS1 deficiency induces endoplasmic reticulum ( ER ) stress , leading to inositol 1 , 4 , 5-trisphosphate receptor ( IP3R ) dysfunction and disturbed cytosolic Ca2+ homeostasis , which , in turn , alters mitochondrial dynamics . Importantly , ER stress , impaired Ca2+ homeostasis , altered mitochondrial dynamics , and delayed neuronal development are causatively related events because interventions at all these levels improved the downstream processes . Our data shed light on the mechanisms of neuronal abnormalities in Wolfram syndrome and point out potential therapeutic targets . This work may have broader implications for understanding the role of mitochondrial dynamics in neuropsychiatric diseases .
Wolfram syndrome ( WS ) is a genetic disorder characterized by diabetes insipidus , diabetes mellitus , optical atrophy , and deafness ( DIDMOAD ) and brain atrophy that results in death in middle adulthood , typically due to brainstem atrophy-induced respiratory failure [1] . About 60% of patients with WS develop a neurological or psychiatric disorder , including psychosis , episodes of severe depression , and impulsive and aggressive behaviour . Importantly , brain abnormalities occur at the earliest stage of clinical symptoms , suggesting that WS has a pronounced impact on early brain development [2] . The majority of WS cases are related to mutations in the gene Wolfram syndrome 1 ( wolframin , WFS1 ) , which encodes a protein localized in the endoplasmic reticulum ( ER ) membrane . A number of studies have pointed out the involvement of WFS1 in Ca2+ homeostasis and ER stress regulation [3–5] . It has been suggested that the ER stress plays a causative role in WS . At the same time , the clinical symptoms of WS resemble mitochondrial disease symptoms , such as deafness , optic atrophy , and psychiatric disorders . Moreover , the affected tissues and organs in WS have a high metabolic demand , and most of the clinical manifestations of WS are consistent with an energy metabolism defect . Therefore , another hypothesis has been forwarded that WS is caused by mitochondrial dysfunction [6 , 7] . This hypothesis is indirectly supported by more recent findings that another causative gene , CISD2 , identified in patients with type 2 WS , is associated with mitochondrial abnormalities and activation of mitophagy [8 , 9] . Importantly , these hypotheses are not mutually exclusive , because ER stress is also capable of impairing mitochondrial function [10] . This rationale further supports the idea that mitochondrial disorders are involved in the pathogenesis of WS . In the present work , using primary neuronal cultures , we show that WFS1 downregulation leads to marked impairment of mitochondrial dynamics , which , in turn , inhibits neuronal development . We demonstrate that WFS1 deficiency triggers an ER stress associated with inositol 1 , 4 , 5-trisphosphate receptor ( IP3R ) dysfunction , leading to altered cell calcium homeostasis . The latter , in turn , is involved in the dysregulation of mitochondrial dynamics ( mitophagy and fusion-fission cycle ) in neurons . These results shed new light on the mechanisms of neuronal abnormalities in WS and point out potentially new therapeutic targets .
Mitochondrial fusion and fission dynamics were measured using photoconvertible mitochondrially targeted Kikume Green-Red , which enables the quantification of fusion events between green- and red-emitting mitochondria ( S1 Video , Fig 1A ) . There was a significant , 3-fold decrease in the number of fusion events in Wfs1 shRNA-treated neurons ( efficiency of Wfs1 shRNA is demonstrated in S1 Fig ) compared with scrambled shRNA-treated controls ( from 0 . 029 ± 0 . 001 to 0 . 010 ± 0 . 001 fusion/mito/min , respectively , n = 80 neurons , p < 0 . 0001; see also Fig 1B , which shows a representative experiment ) . Also , the mitochondrial fission rate was decreased in Wfs1 shRNA-treated neurons ( from 0 . 027 ± 0 . 001 to 0 . 010 ± 0 . 001 fission/mito/min , n = 80 neurons , p < 0 . 0001 ) , suggesting that WFS1 deficiency greatly prolongs the fusion–fission cycle . These changes in the fusion–fission cycle were associated with a 20% decrease in mitochondrial length ( Fig 1C ) . Overexpression of human shRNA-insensitive wild-type ( wt ) WFS1 but not P724L mutant ( which occurs in WS ) restored the mitochondrial fusion dynamics and length in the Wfs1 shRNA-treated group ( Fig 1B and 1C ) . The inhibition of mitochondrial fusion was also observed in cerebellar granule neurons in which the fusion rate decreased from 0 . 036 ± 0 . 006 fusion/mito/min in control to 0 . 008 ± 0 . 003 fusion/mito/min in WFS1-deficient neurons ( n = 16 neurons , p = 0 . 001 ) . We also quantified mitochondrial fusion and fission dynamics in neurons isolated from Wfs1-/- and Wfs1+/+ mice . Of significance , a 2 . 5-fold decrease in the number of fusion events in neurons from Wfs1-/- compared with those from Wfs1+/+ mice was observed ( Fig 1D ) . Again , this decrease was associated with an approximate 25% decrease in mitochondrial length ( Fig 1E ) . Moreover , Wfs1 shRNA suppression of fusion rate and induction of mitochondrial shortening in Wfs1+/+ neurons had no effect in Wfs1-/- neurons , thus showing its specificity . In contrast , overexpression of shRNA-insensitive WFS1 restored the mitochondrial fusion dynamics and length in Wfs1-/- neurons while having relatively little effect in Wfs1+/+ neurons ( Fig 1D and 1E ) . No difference , however , was observed between Wfs1-/- and Wfs1+/+ brains in mRNA expression of the main fusion and fission proteins ( S2 Fig ) . To measure mitophagy , we expressed the mitochondrially targeted pH-dependent protein Keima , the excitation spectrum of which shifts ( green to red in Fig 1F ) when mitochondria are delivered to acidic lysosomes . The number of autolysosomes containing mitochondria was increased both in wt neurons transfected with Wfs1 shRNA and in neurons isolated from Wfs1-deficient animals ( Fig 1G and 1H ) . Also , the number of mitochondria-containing autophagosomes ( LC3-positive dots co-localizing with a mitochondrial marker; S3 Fig ) was significantly increased in neuronal bodies of Wfs1 shRNA-treated neurons ( 13 . 8 ± 1 . 1 versus 8 . 8 ± 0 . 8 in scrambled shRNA-transfected neurons , p = 0 . 0004 , n = 50 neurons ) . Overexpression of human shRNA-insensitive wt WFS1 in the Wfs1 shRNA-treated group reduced the mitophagy close to control level ( Fig 1G ) . Note that Wfs1 silencing enhances LC3-I conversion to LC3-II ( S4 Fig ) , suggesting increased autophagy . Together , the LC3 and Keima assays demonstrate that WFS1 deficiency increases mitochondrial removal ( mitophagic flux ) rather than inhibiting the formation of autolysosomes containing mitochondria . Importantly , WFS1-deficient neurons showed fewer mitochondria in axons . Analysis of mitochondrial density in neurites revealed an approximate 30% decrease in mitochondrial mass both for Wfs1 shRNA-treated neurons and for neurons isolated from Wfs1-/- mice ( Fig 1I–1K ) . This effect was phenocopied by Wfs1 shRNA in Wfs1+/+ neurons and rescued by wt WFS1 overexpression in Wfs1-/- neurons . Mitochondrial trafficking was also disturbed in WFS1-deficient neurons . Mitochondria in these neurons showed a decrease in velocity in both antero- and retrograde motion ( Table 1 , S5 Fig ) . Mitochondria from WFS1-deficient neurons also changed their direction during motion more often and made more individual runs , whereas the length of these runs was shorter . These data explain the oscillatory-like movement we frequently observed in WFS1-deficient neurons ( S2 and S3 Videos ) . We also measured the contact rate of mitochondria as an indirect parameter of mitochondrial movement . This parameter was also significantly reduced in neurons isolated from Wfs1-deficient mice and restored by WFS1 overexpression ( S6 Fig ) . We performed a quantitative analysis of mitochondrial membrane potential using the ratiometric mitochondrial membrane potential-sensitive fluorescent probe JC-10 ( emitting light from 525 nm to 590 nm depending on mitochondrial membrane potential ) . Neurons were first transfected with Wfs1 siRNA using the N-TER nanoparticle siRNA transfection system to ensure >70% transfection efficiency . The results obtained demonstrated a 10% decrease in red to green fluorescence ratio , suggesting a slight depolarisation in the Wfs1 siRNA group ( Figs 2A and S7 ) . Interestingly , an increased number of polarised , tetramethylrhodamine ethyl ester ( TMRE ) -positive mitochondria were observed inside autophagosomes in WFS1-deficient neurons ( S8 Fig ) ; this suggests that WFS1 deficiency may induce mitophagy of active , polarised mitochondria . We also performed indirect reactive oxygen species ( ROS ) measurements using an NRF2 reporter gene assay , which showed no difference between the scrambled and Wfs1 shRNA-transfected neurons ( Fig 2B ) . Next , we estimated cell ATP levels using the fluorescence resonance energy transfer ( FRET ) -based ATP sensor ATeam , which employs the epsilon subunit of a bacterial F0F1-ATPase . Control experiments with deoxyglucose/oligomycin or glutamate reduced neuronal ATP levels and demonstrated the expected decline in FRET signal , showing the validity of our approach ( Fig 2C ) . The results obtained ( Fig 2D ) show that the cellular ATP level is reduced in WFS1-deficient neurons . It has been previously demonstrated that WFS1 deficiency leads to ER stress in different rodent and human cell lines [5] , WFS1-deficient β-cells [11] , and WFS1-deficient mouse retinas [12] . We checked whether the suppression of WFS1 in neurons also induced ER stress . Using reporter constructs for ATF6 , IRE1-XBP1 , and PERK-ATF4 pathways , we demonstrate ( Fig 3A ) that WFS1 suppression activates luciferase reporter constructs with a promoter containing ATF6 and ATF4 binding sites , but not the XBP1 splicing reporter . This WFS1 deficiency-induced activation , however , was weak when compared with that induced by overexpression of ATF6 , ATF4 , and IRE1 ( Fig 3B ) . To test whether the effect of WFS1 deficiency on mitochondrial dynamics is associated with ER stress , we mitigated ER stress by overexpressing the ER chaperon HSPA5 . Fig 3C and 3D demonstrate that wt HSPA5 and its ATPase-deficient mutant ( T37G ) , but not its protein binding-defective mutant ( P495L ) , restore the Wfs1 shRNA-suppressed fusion rate . Also , wt HSPA5 partially suppressed WFS1 deficiency-induced mitophagy ( Fig 3E ) . Importantly , activation of major ER stress-response pathways by overexpressing ATF6 , ATF4 , or IRE1 neither inhibited mitochondrial fusion nor induced mitochondrial shortening or mitophagy ( Fig 3F–3H ) . Furthermore , ATF4 and ATF6 shRNAs were not able to restore mitochondrial fusion rate or decrease mitophagy in WFS1-deficient neurons ( Fig 3I–3L ) . These data suggest that although WFS1 deficiency-induced mild ER stress is followed by disturbed mitochondrial dynamics , the latter is not mediated by these major ER stress-response pathways . An earlier report demonstrated that ER stress-impaired IP3R-mediated Ca2+ release from the ER [13] . It is noteworthy to mention that IP3R rather than ryanodine receptors are primarily responsible for Ca2+ release in cortical neurons ( S9A Fig ) . To test whether IP3R-mediated Ca2+ homeostasis is impaired in WFS1-deficient neurons , we loaded neurons with both the Ca2+ sensor Fluo-4 and a membrane-permeant caged derivative of IP3 . IP3 uncaging-induced Ca2+ release from the ER to cytosol was notably decreased in WFS1-deficient neurons ( Fig 4A ) . Similarly , the selective group I metabotropic glutamate receptor agonist dihydroxyphenylglycine ( DHPG ) , which stimulates phospholipase C and promotes endogenous IP3 formation , induced diminished cytosolic Ca2+ transients in WFS1-deficient neurons ( Fig 4B ) . No difference was observed in basal ER [Ca2+] ( S10 Fig ) or in maximal ER Ca2+ uptake capacity ( S9B Fig ) , suggesting that the decreased IP3-dependent Ca2+ release was not due to reduced ER Ca2+ levels . Also , no difference was observed in the cytosolic Ca2+ transients elicited by an inhibitor of endoplasmic reticulum Ca2+ ATPase , 30 μM cyclopiazonic acid ( CPA ) , suggesting that there is no difference in releasable ER Ca2+ between control and Wfs1 silenced neurons ( S9C and S10D Figs ) . It has been proposed that the IP3 receptor is involved in Ca2+-induced Ca2+ release from the ER ( for review , see [14] ) . Thus , it is reasonable to suggest that impaired IP3R function would affect Ca2+ release from the ER during neuronal depolarisation . Indeed , by following changes in cytosolic Fluo-4 fluorescence , we observed that cytosolic Ca2+ transients in neurons in response to glutamate or KCl were up to 2-fold lower in WFS1-deficient neurons when compared with control groups ( Fig 4C and 4D ) . We next tested whether decreased IP3-dependent Ca2+ release is associated with altered basal cytosolic [Ca2+] levels . We used a FRET-based Ca2+ sensor to enable measurement of cytosolic [Ca2+] before and after KCl treatment . The data obtained demonstrated increased basal [Ca2+] in WFS1-deficient neurons , whereas stimulation led to lower maximal [Ca2+] ( Fig 4E–4G ) . Accordingly , the amplitude of Ca2+ transient was significantly decreased ( Fig 4H ) . Similarly , a decrease was observed when aequorin-emitted bioluminescence was used to quantify the maximal [Ca2+] after stimulation ( Fig 4I ) . Furthermore , pre-treatment with Araguspongin B , an inhibitor of IP3-dependent Ca2+ release , suppressed KCl-induced cytosolic Ca2+ transients in wt neurons ( Fig 5A ) , whereas overexpression of the active fragment of IP3R restored the cytosolic Ca2+ transients in WFS1-deficient neurons ( Fig 5D ) . These findings suggest that WFS1 deficiency induces lower IP3R-mediated Ca2+ release . Next , we checked whether the disturbed cytosolic Ca2+ homeostasis could be responsible for the impaired mitochondrial dynamics observed in WFS1-deficient neurons . Indeed , IP3R inhibition by Araguspongin B suppressed the mitochondrial fusion rate and initiated mitophagy in wt neurons ( Fig 5B and 5C ) . In contrast , overexpression of the active fragment of IP3R corrected the WFS1 deficiency-induced changes in mitochondrial dynamics ( Fig 5E and 5F ) . The latter suggests a direct link between IP3R-mediated ER calcium release and mitochondrial dynamics . Furthermore , treatment of WFS1-deficient neurons with the L-type Ca2+ channel activator , Bay K 8644 , restored cytosolic Ca2+ transients ( Fig 5G ) , restored the fusion rate , and inhibited mitophagy ( Fig 5H and 5I ) . A similar rescue was obtained by overexpressing plasma membrane ORAI calcium release-activated calcium modulator 1 ( ORAI1 ) , which also corrected cytosolic Ca2+ transients , restored the fusion rate , and suppressed mitophagy in WFS1-deficient neurons ( Fig 5J–5L ) . These experiments suggest that disturbed cytosolic Ca2+ homeostasis , rather than ER Ca2+ specifically , or direct ER-mitochondria Ca2+ channelling , is responsible for impaired mitochondrial dynamics in WFS1-deficient neurons . Because WS has been shown to be associated with both neurodegeneration and impaired early brain development [2] , we further aimed to test whether the alterations in Ca2+ homeostasis and mitochondrial dynamics affect neuronal development or/and survival . WFS1 deficiency delayed the development of cortical neurons markedly ( Fig 6A and 6B; see S1 Table for original data ) . The longest axon and the axonal tree were significantly shorter , and the number of axonal tips was lower in developing DIV2-DIV4 Wfs1 shRNA-transfected neurons ( Fig 6C–6F ) . However , in relatively mature neurons , at DIV6 , the axonal length and branching was similar to control . Fig 6G demonstrates that the survival of WFS1-deficient neurons is also impaired; transfection of Wfs1 shRNA led to relatively slight but significant loss of neurons . Interestingly , WFS1 suppression decreased significantly the density of synapses when measured at DIV19 but not at earlier stages ( Fig 6H and 6I ) . To examine whether the compromised development and survival we observed in vitro have relevance in vivo , we further conducted ex vivo magnetic resonance imaging of the brains of 1-y-old Wfs1-deficient male mice . Volumetric analysis did not demonstrate a significant change in total cerebral volume , but there was a marked reduction of the optic nerve and brain stem volumes in Wfs1 deficient mice ( Fig 7A–7G ) . A slight decrease was also observed in cortical area at the level of striatum ( Fig 7H ) . Thus , these results suggest that WFS1 is indispensable for appropriate neuronal development , morphology , and survival both in vitro and in vivo . We next tested the possibility that delayed neuronal development in WFS1 deficiency is a consequence of IP3R dysfunction and/or impaired mitochondrial dynamics . If this hypothesis is correct , it could open the possibility of improving neuronal development by restoring cytosolic Ca2+ homeostasis or mitochondrial dynamics . Indeed , overexpression of IP3R improved Ca2+ homeostasis , protected against WFS1 deficiency-induced developmental delay and also partially restored axonal growth ( Fig 8A–8C ) . Importantly , this overexpression did not suppress ER stress ( S11 Fig ) , thus suggesting that IP3R-mediated Ca2+ disturbances rather than ER stress per se are responsible for the neuronal development delay . Furthermore , overexpression of ORAI1 normalized cytosolic Ca2+ homeostasis and also protected neurons against the development delay ( S12 Fig ) . Finally , we tested whether specific suppression of mitophagy in WFS1 deficiency by co-expression of Wfs1 shRNA with Pink1 or Parkin shRNAs could rescue the neuronal development . Both shRNAs suppressed effectively the WFS1 deficiency-induced mitophagy and restored mitochondrial density ( Fig 8D , 8H , 8E and 8I ) , demonstrating that WFS1 deficiency activates selective Pink1-Parkin-dependent mitophagy . The latter suggestion is supported by the finding that Parkin translocation to mitochondria was higher in Wfs1 shRNA-expressing PC6 cells , PC12 cells ( S13 Fig ) , and in neurons ( S14 Fig ) . Moreover , Pink1 and Parkin shRNAs restored the fusion rate and contact rate ( an indirect parameter for mitochondrial movement; Fig 8F , 8J , 8G and 8K ) , suggesting the relevance of mitophagy in these processes . Most importantly , suppression of mitophagy by expressing Pink1 or Parkin shRNAs accelerated development and restored the axonal growth in WFS1-deficient neurons ( Fig 8L–8Q ) . At the same time , Parkin and Pink1 shRNAs neither suppressed WFS1 deficiency-induced ER stress nor restored Ca2+ transients ( S15 Fig ) , suggesting that the impairment of mitochondrial dynamics is a downstream event relative to the Ca2+ homeostasis disturbances and that overactivated mitophagy is the primary reason for delayed neuronal development in WFS1-deficient neurons . We also tested whether the inhibition of mitochondrial fission proteins could protect mitochondria in WFS1-deficient neurons and rescue the neurons from the developmental delay . Treatment with negative dominant DRP1 ( nd DRP1 ) reversed the negative effects of WFS1 deficiency on fusion and density loss and restored normal development ( S16A–S16C Fig ) . nd DRP1 also protected against the inhibition of mitochondrial fusion induced by the ER stressor Brefeldin A ( S16D Fig ) . Brefeldin A itself showed too strong a negative effect on neuronal survival , making it impossible to estimate neuronal development .
Our results uncover a chain of causal links relating ER stress , cytosolic Ca2+ disturbances , impaired mitochondrial dynamics , and delayed neuronal development in WFS1-deficient neurons . We demonstrate that WFS1 deficiency induces ER stress in neurons ( as has already been shown in other cell types [5 , 15] ) , affecting the IP3R receptor . This was associated with higher cytosolic Ca2+ at resting conditions ( which is consistent with previously reported elevated basal cytosolic [Ca2+] in Wfs1 deficient iPS cells [16] ) but lower maximal [Ca2+] under stimulated conditions , suggesting reduced amplitude of IP3R-mediated Ca2+ release in WFS1-deficient neurons . We show that the amplitude of IP3R-mediated Ca2+ release induced by photolysis of caged IP3 or by activating endogenous IP3 production by the metabotropic glutamate receptor agonist DHPG was significantly lower in WFS1-deficient neurons . The exact mechanism of this WFS1-dependent IP3R dysfunction is not clear; however , it has been previously shown that ER stress induced IP3R inhibition by impairing the IP3R-HSPA5 interaction [13] . We further demonstrate that altered Ca2+ homeostasis disturbs mitochondrial dynamics . Mitochondria in WFS1-deficient neurons do not move properly , they do not fuse and split apart as frequently as their wt counterparts , and they undergo mitophagy more frequently . Overexpression of the active IP3R fragment restores IP3R-mediated Ca2+ release and corrects all perturbations in mitochondrial dynamics , suggesting that these events are causally linked . Pharmacological inhibition of IP3R by Araguspongin B phenocopied the effects of WFS1 deficiency , confirming the connection between reduced ER Ca2+ release and impaired mitochondrial dynamics . Importantly , we were able to correct mitochondrial dynamics by activating store-operated calcium entry or by activating L-type Ca2+ channels pharmacologically , linking the lower ER Ca2+ release with impaired mitochondrial dynamics . Potentially , there are several ways how ER Ca2+ release could influence mitochondrial dynamics . Lowered levels of ER Ca2+ release could directly activate/deactivate mitochondrial and/or cytoskeleton proteins involved in mitochondrial dynamics . One may suggest that Ca2+ affects activity or expression of these proteins through the calcium/calmodulin ( CaM ) kinase signalling cascade , which may not be sufficiently activated in WFS1 deficiency . This question deserves specific further study . We also demonstrate that the mechanism linking WFS1 deficiency-related ER stress with impaired mitochondrial dynamics involves two Parkinson’s disease-related proteins , PINK1 and Parkin . Both Pink1 and Parkin shRNAs supressed WFS1 deficiency-induced mitophagy back to control levels . These data suggest that WFS1 deficiency may activate the PINK1 and Parkin pathway ( supported by our finding demonstrating increased Parkin translocation to mitochondria under basal conditions ) , which has been shown to inhibit mitochondrial movement [17] and fusion-fission dynamics [18–20] and induce mitophagy [21–24] . Importantly , Pink1 and Parkin shRNAs also restored mitochondrial fusion–fission dynamics and trafficking , suggesting activation of PINK1-Parkin pathway to be the primary event leading to impaired trafficking and fusion rate as well as to mitophagy . The result that PINK1 or Parkin silencing ( which improves mitochondrial dynamics in WFS1 deficient neurons ) does not correct ER stress Ca2+ responses in WFS1 deficiency suggests that regulation of mitochondrial dynamics by the PINK1-Parkin pathway is downstream to cytosolic Ca2+ homeostasis . In principle , there are two potential explanations for how PINK1-Parkin pathway could be involved . First , mitochondrial depolarization could lead to PINK1 accumulation in the mitochondrial outer membrane and Parkin translocation to mitochondria , inhibiting mitochondrial fusion and trafficking and inducing mitophagy . This explanation could be supported by our finding that mitochondria in WFS1-deficient neurons were slightly depolarised . Another explanation would be that overactivation of the PINK1-Parkin pathway occurs independently of mitochondrial membrane potential , leading to the removal of healthy and polarised mitochondria . We earlier observed a similar phenomenon in mutant alpha-synuclein expressing neurons where PINK1-Parkin-dependent mitophagy started to eliminate polarised mitochondria [25] . It cannot be excluded that both of these explanations are also valid for WFS1-deficient neurons . Slight mitochondrial depolarisation may increase the rate of mitochondrial removal , and PINK1-Parkin dependent mitophagy could start to eliminate functional or at least partly functional mitochondria . This excessive mitochondrial removal should then lead to decreased mitochondrial density and ATP production , both of which were observed in WFS1-deficient neurons and compromise the bioenergetic status of cells . Compared with the majority of other cell types , in which mitochondrial turnover is high , mitochondrial turnover is relatively low in neurons . It might therefore be suggested that neurons cannot afford to lose mitochondria at a high rate , as it would lead to energy deficits . Instead , it might be energetically more favourable for neurons to keep “partially defective” mitochondria than to consume them through mitophagy; in other words , “partially defective” mitochondria are the lesser evil . In contrast , under pathological conditions associated with increased levels of autophagy and mitophagy , excessive and unwanted mitochondrial clearance would lead to bioenergetic deficits harmful to neurons ( in our case , an increased number of partially defective mitochondria and increased removal of these partially defective mitochondria , leading to reduced mitochondrial mass ) . It is likely that this event is not limited to Wfs1-deficient neurons but might also be observed in the future in other neurodegenerative conditions . There is also some limited evidence in the literature that WFS1 is associated with Parkinson pathways . Shadrina et al . [26] demonstrated that the synonymous polymorphism C1645T in the WFS1 gene increases the risk of Parkinson's disease in Russian patients . Kõks et al . [27] demonstrated recently that WFS1 silencing in HEK cells primarily affected the expression of genes belonging to the Parkinson’s signalling ingenuity canonical pathway . Moreover , WFS1-deficient mice demonstrate impaired function of the dopaminergic system [28] . Another important discovery is that WFS1 deficiency delays neuronal development and impairs neuronal survival in primary neuronal culture . Ex vivo magnetic resonance imaging of the brains of Wfs1-/- mice also demonstrated clear atrophy and/or degeneration of the brain stem , which is the main structure atrophied in Wolfram syndrome patients and the cause of death due to respiratory failure [1] . This is also consistent with an earlier clinical study suggesting that WFS1 had a pronounced impact on early brain development [2] . Compared with healthy and type 1 diabetic control groups , a cohort of young WS patients at relatively early stages of disease showed smaller intracranial volumes and preferentially affected grey matter volume and white matter microstructural integrity . According to our data , the link between WFS1 deficiency and delayed neuronal development appears to be mediated by impaired mitochondrial dynamics , because suppression of the PINK1-Parkin pathway also corrected the development delay . Our results do not allow us to elucidate the exact mechanism by which disturbed mitochondrial dynamics delays neuronal development; however , some putative mechanisms could be proposed . Impaired mitochondrial trafficking in WFS1-deficient neurons might negatively affect the delivery of energy-producing mitochondria to the sites where energy is most needed . For example , inhibition of mitochondrial transport results in the loss of mitochondria from peripheral nerve terminals that may reduce local ATP supply and affect ATP-dependent processes . Besides , any serious disturbances in mitochondrial fusion–fission dynamics may impair the maintenance of mitochondrial function and further compromise neuronal energy requirements . Disruption of mitochondrial fusion results in mitochondrial dysfunction and loss of respiratory capacity ( for review , see [29] ) . Finally , excessive mitophagy that decreases mitochondrial mass will also affect the capacity of total energy production in neurons . Notably , neuronal growth is associated with increased energy demand and may be slowed by energy deficits . Our discovery that WFS1 deficiency-elicited perturbations in Ca2+ homeostasis leads to disturbed mitochondrial dynamics and impaired neuronal development may help us to understand the pathophysiology of some psychiatric disorders . Indeed , WS has been associated with psychiatric pathologies ( for review , see [30] ) , such as severe depression , psychosis , dementia , impulsive-aggressive behaviour leading to suicide attempts , and frequent hospitalization [31] . Heterozygous carriers of mutant WFS1 , who are estimated to be as high as 1% of the general population , may also be at increased risk for mood disorders . Swift et al . [32] suggested that heterozygous WS carriers are 26-fold more likely to require psychiatric hospitalization compared with non-carriers , and these heterozygotes may constitute approximately 25% of all individuals hospitalized with depression and suicide attempts . These findings were confirmed in several further papers [33–35] , although some reports failed to find an association [36–38] . This discrepancy is likely related to differences in cohorts of patients and requires further investigation . However , in general , our data are consistent with a growing body of evidence suggesting that impaired mitochondrial function ( including mitochondrial dynamics ) may lead to a disruption of normal neural plasticity and reduced cellular resilience , which may , in turn , promote the development of mood and psychotic disorders . In conclusion , our data suggest a causal relationship between ER stress , cytosolic Ca2+ disturbances , impaired mitochondrial dynamics , and delayed neuronal development in WFS1-deficient neurons . This mechanism sheds new light on the development of neuronal abnormalities in Wolfram syndrome and points out potential therapeutic targets . Moreover , our results unravel two rather unexpected links having impact beyond the relatively rare Wolfram syndrome . Firstly , relatively mild ER stress/impaired ER Ca2+ release could seriously disturb mitochondrial dynamics , thus providing an explanation as to why alterations at the ER level could lead to a mitochondrial phenotype . Secondly , impaired mitochondrial dynamics could affect neuronal development , suggesting that proper mitochondrial dynamics might be crucial for neurodevelopment . Since alterations in WFS1 function seem to take place in different neurological disorders [30 , 32] , our work may also have rather broad implications for understanding the role of mitochondrial dynamics in neuropsychiatric diseases .
Plasmids expressing scrambled shRNA or shRNA against rat Wfs1 ( KR46208N ) , rat Atf6 ( KR51427H ) , rat Atf4 ( R42749N ) , rat Parkin ( KR50238N ) , and rat Pink1 ( KR55105N ) were from SABiosciences . shRNAs against Parkin and PINK1 have been validated by us earlier [20] . shRNAs against ATF6 and ATF4 supressed the expression of respective mRNAs by 74% and 68% . Plasmids expressing mitochondrial DsRed2 ( 632421 ) and EGFP ( 6085–1 ) were from Clontech . Mito-Keima was from Amalgaam ( AM-V0251 ) , and mito-KikGR1 was constructed as described earlier [39] . ATeam ( 51958 ) , ATF6 ( 11975 ) , ATF6-GL3 ( 11976 ) , ATF4 ( 26114 ) , ATF4-luc ( 21850 ) , WFS1 wt ( 13011 ) , WFS1 P724L ( 13012 ) , IRE1α ( 13009 ) , D1ER ( 36325 ) , D3cpv ( 36323 ) , DRP1 K38A , EGFP-LC3 ( 24920 ) , HSPA5 wt ( 27164 ) , HSPA5 T37G ( 27165 ) , HSPA5 P495L ( 27166 ) , NRF2 ( 21555 ) , ORAI1 ( 21638 ) , PSD-95 ( 15463 ) , and pAAV-hSyn-DsRedExpress ( 22907 ) were obtained from Addgene ( Cambridge , MA ) . The IP3R1 channel fragment ( MmCD00312368 ) was from PlasmID , pRL-CMV ( E2261 ) was from Promega Co . , and mKate2-mito ( FP187 ) was from Evrogen . XBP1ΔDBD-LUC was a kind gift from Dr . T . Iwawaki , YFP-Parkin from Dr . R . Youle , cytosolic aequorin from Dr . R . Rizzuto , DRP1 K38A from Dr . G . Szabadkai , pGL3-rNQO1 Dr . J . Alam , and PINK1 from Dr . E . Deas . DHPG ( 0805 ) and Bay K 8644 ( 1544 ) were from Tocris Bioscience , Brefeldin A ( B6542 ) and Cyclopiazonic acid ( C1530 ) were from Sigma-Aldrich , and Araguspongin B was from Cayman Chemical ( 10006797 ) . All fluorescence dyes and culture media were from Life Technologies . Primary rat neuronal cultures were prepared from less than 1-d-old neonatal Wistar rats as described earlier [39] . Briefly , cortices were dissected in ice-cold Krebs–Ringer solution containing 0 . 3% BSA and then trypsinised in 0 . 8% trypsin for 10 min at 37°C . The cells were then triturated in a 0 . 008% DNase solution containing 0 . 05% soybean trypsin inhibitor . Cells were resuspended in Basal Medium Eagle with Earle's salts ( BME ) containing 10% heat-inactivated fetal bovine serum ( FBS ) , 25 mM KCl , 2 mM glutamine , and 100 μg/ml gentamicin , and then plated onto 35-mm glass-bottomed dishes ( MatTek , MA ) , which were pre-coated with poly-L-lysine , at a density of 106 cells per dish in 2 ml of cell suspension . After incubating for 3 h , the medium was changed to Neurobasal-A medium containing B-27 supplement , 2 mM GlutaMAX-I , and 100 μg/ml gentamicin . To prepare primary cultures of cerebellar granule cells , the cerebella from 8-d-old Wistar rats were dissociated by trypsinising in 0 . 25% trypsin at 35°C for 15 min , followed by trituration in a 0 . 004% DNase solution containing 0 . 05% soybean trypsin inhibitor . Cells were resuspended in BME containing 10% FBS , 25 mM KCl , 2 mM glutamine , and 100 μg/ml gentamicin . Neurons were plated onto 35 mm glass-bottomed dishes that were pre-coated with poly-L-lysine at a density of 1 . 3 × 106 cells/ml . 10 μM cytosine arabinoside was added 24 h after plating to prevent the proliferation of glial cells . Primary cortical neurons were isolated using the same protocol from 1-d-old wt and Wfs1-deficient mice [40] obtained from mating of background-matched wt and Wfs1-deficient mice , respectively . The permissions for the animal studies were given to E . Vasar ( No . 39 and 29 ) and D . Safiulina ( No . 51 ) by the Estonian National Board of Animal Experiments in accordance with the European Communities Directive of 24 November 1986 ( 86/609/EEC ) . PC12 or PC6 cells were grown in RPMI-1640 medium supplemented with 10% horse serum and 5% FBS on collagen IV-coated 100-mm plastic dishes or on 35-mm glass-bottomed dishes . All culture media and supplements were obtained from Invitrogen . Neurons were transfected at DIV2 ( with exception of neuronal maturation and axonal growth experiments ) . For transfection of cells growing on glass-bottomed dishes , the conditioned medium was replaced with 100 μl Opti-MEM I medium containing 2% Lipofectamine 2000 and 1 to 2 μg of total DNA with an equal amount of each different plasmid . The dishes were incubated for 3 to 4 h , after which fresh medium was added . For biochemical analyses , the cells were transfected in 100-mm plastic dishes as described above , except that the total volume of the transfection mixture was increased with proportionally adjusted Lipofectamine 2000 and DNA . For some experiments with shRNA-expressing plasmids that also contained a neomycin resistance gene ( shRNA efficiency testing and Parkin translocation ) , the PC12 cell medium was supplemented with 200 μg/ml G418 for 6–7 d . Note that despite relatively low transfection efficiency in neurons , the transfected plasmids were mostly localised to the same cells . When neurons were transfected with plasmids encoding mitochondrial CFP , mitochondrial YFP , and mitochondrial mKate2 , 93 ± 1% transfected cells expressed all three markers , 5 ± 1% expressed two markers , and 2 ± 0 . 4% expressed one marker ( n = 4 dishes; 100 cells were analysed per dish; S17 Fig ) . For mitochondrial fusion rate analysis , cortical neuronal cultures transfected with mito-KikGR1 plasmid and plasmids of interest as described earlier [39] and examined at DIV 7–8 . A laser scanning confocal microscope ( LSM 510 Duo , Carl Zeiss Microscopy GmbH ) equipped with a LCI Plan-Neofluar 63×/1 . 3 water immersion DIC M27 objective was used . The temperature was maintained at 37°C using a climate chamber . For fusion acquisition , mito-KikGR1 was illuminated with a 488-nm argon laser line to visualize the intense green mitochondrial staining . Selected mitochondria were then photoconverted to red using a 405-nm diode laser and illuminated using a 561 nm DPSS laser . The images were taken at 10-s intervals for 10 min , the fate of all activated mitochondria was followed throughout the time-lapse , and the fusion and fission events were recorded . For whole-cell mitochondrial density measurements , the neurons were transfected with GFP , mitochondrial pDsRed2 , scrambled shRNA or shRNA , and plasmids of interest . At DIV 8 , the entire axon and dendrites from randomly selected neurons were visualised using a laser scanning confocal microscope . Neurons were reconstructed using Neurolucida and LSM5 software , and mitochondrial density was analysed . Mitochondrial length measurements were performed as described previously [39] . Primary cortical neurons transfected with mitochondrially targeted Keima plus scrambled shRNA , and plasmids of interest were studied at DIV 7–8 . The excitation spectrum of Keima shifts from 440 to 586 nm when mitochondria are delivered to acidic lysosomes , which enables quantification of mitophagy . Images were acquired by a laser scanning confocal microscope using the laser lines 458 nm ( green , mitochondria at neutral pH ) and 561 nm ( red , mitochondria under acidic pH ) at 5–6 d after transfection , and red dots were counted blindly . In another set of experiments , neurons were transfected with pEGFP-LC3 , mKate2-mito , and scrambled shRNA or Wfs1 shRNA . The co-localization of EGFP-LC3 dots and mitochondrial mKate2-mito was analysed . Neurons expressing the ATP sensor ATeam and scrambled or Wfs1 shRNA were excited using a 458 nm line ( 10% ) of Ar-laser , the CFP emission was acquired at 465–500 nm and the FRET signal at 520–570 nm . The ratio of the FRET/CFP fluorescence intensity was calculated from the signal coming from the cytosol . For membrane potential measurements , primary neurons ( plated at lower density , 2 . 5 x 105 cells/ml ) were transfected with 20 nM validated siRNA against Wfs1 ( Sigma-Aldrich: SASI_Rn02_00265296 Rat NM_031823; Wfs1 siRNA suppressed 80 ± 1% of endogenous WFS1 expression in primary cortical cells as estimated by RT-PCR , n = 3 ) using the N-TER Nanoparticle siRNA Transfection System ( Sigma-Aldrich ) . Briefly , a mixture of target and scrambled siRNA ( 20 nM ) diluted in siRNA buffer and NTER transfection reagent diluted in ddH20 was preincubated at RT for 20 min . Growth medium was then removed and replaced with Opti-MEM I containing the target or scrambled siRNA mixture . After a 3-h incubation at 37°C , Opti-MEM I was changed to Neurobasal-A medium containing B-27 supplement , 2 mM GlutaMAX-I , and 100 μg/ml gentamicin . After transfection , the cells were incubated for 72–96 h in a humidified 5% CO2/95% air incubator at 37°C . The N-TER Nanoparticle siRNA Transfection System was relatively non-toxic , yielding a survival rate of 82 ± 4% ( five dishes and five dishes in vehicle treated group ) at 16 h after transfection ( estimated with LIVE/DEAD Viability/Cytotoxicity Kit , for mammalian cells [Invitrogen] ) . For JC-10 loading , siRNA-transfected dishes were kept in 10 μM JC-10 dissolved in culture media and incubated at 37°C for 20 min . After dye-loading , the cells were kept in Krebs-Ringer solution supplemented with 1mM Ca2+ and visualized using a laser scanning confocal microscope equipped with a LCI Plan-Neofluar 63×/1 . 3 water immersion DIC M27 objective . Dishes were then treated with 5 μM FCCP to obtain background values . For visualizing mitochondria with preserved membrane potential in autophagosomes , the neurons were transfected with LC3-EGFP and Wfs1 shRNA . On the fourth day after transfection , the cells were stained with the mitochondrial membrane potential sensitive dye tetramethylrhodamine ethyl ester ( TMRE ) at a concentration of 50 nM in complete NeurobasalTM-B medium at 37°C for 30 min and visualized in Krebs-Ringer solution supplemented with 1 mM Ca2+ using confocal microscope equipped with a 100×/oil objective . For Fluo-4 based measurements , neurons were transfected with scrambled shRNA or Wfs1 shRNA , plasmids of interests , and mKate2-mito to visualise transfected cells . Five days later , cells were loaded with 5 μM Fluo-4 AM in Hank’s Balanced Salt Solution ( HBSS with Ca2+ and Mg2+ ) for 30 min at 37°C , followed by 10 min incubation in HBSS without dye to allow complete de-esterification of intracellular AM esters . Fluo-4 AM was excited using a 488-nm argon laser , and emitted fluorescence was quantified using a LSM 510 confocal microscope . Time-lapse images were recorded at 2-s intervals for 1 min before and 2 min after the induction of Ca2+ transients . Cytosolic Ca2+ transients were induced by 25 mM KCl , 200 μM DHPG , 2 mM glutamate , or by the photoactivatable membrane-permeant caged derivative of IP3 . In the latter case , cells were co-loaded with 2 . 5 μM Fluo4-AM and 4 μM Ins ( 1 , 4 , 5 ) P3-PM ( SiChem ) for 90 min at 37°C . Photolysis of caged IP3 by irradiating the individual cells with near-UV laser ( 405 nm ) for 10 s was used to release active IP3 . Fluorescence signals from single transfected neurons identified with mKate2-mito were analysed , and mean changes in fluorescence intensities were calculated . Examples of Fluo-4 time-lapse images in control or Wfs1-siRNA transfected neurons are depicted in S18 Fig . For the FRET-based analysis of cytosolic and ER Ca2+ levels , neurons were transfected with the genetically encoded FRET-based chameleon indicators D3cpv or D1ER , respectively , and scrambled shRNA or Wfs1 shRNA . Transfected neurons were rinsed once and then allowed to equilibrate in HBSS containing Ca2+ and Mg2+ for 20 min . Neurons were excited at 405 nm and emission acquired at 465–510 ( CFP ) and 520–555 ( FRET ) . For bioluminescence-based calcium measurements , neurons were transfected with cytosolic aequorin together with shRNAs and plasmids of interest . Five days later , cells were incubated for 30 min with 3 μM ViviRen Live Cell Substrate ( Promega ) in Krebs buffer containing 1 mM CaCl2 and 0 . 5% BSA at 37°C . Ca2+ uptake was stimulated by 25 mM KCl and experiments were terminated by lysing the cells with 2 . 5% Triton X-100 in Ca2+-rich solution to measure the maximal activity of aequorin . Aequorin luminescence was monitored by Victor X5 Multilabel Plate Reader ( PerkinElmer ) . Primary neurons or PC12 cells growing in 96-well plates were transfected with the desired firefly luciferase reporter plasmid , Renilla luciferase , and plasmids of interest . The luciferase assays were performed 48–96 h later using Dual-Glo Luciferase Assay reagent ( Promega ) according to the manufacturer's instructions . The promoter activities for NRF2 , ATF6 , ATF4 , and IRE firefly luciferase luminescence were normalized to Renilla luciferase signal . For neuronal maturation experiments , cortical neurons were transfected at DIV1 with a plasmid expressing neuron-specific pAAV-hSyn-DsRed1 and scrambled shRNA or Wfs1 shRNA . Live-cell morphology was visually examined using a fluorescence microscope ( Olympus IX70 , 20x/0 . 70 water immersion objective ) on randomly selected fields ( minimum 30 fields per group ) on the indicated days in culture . Neurons were classified into the four subgroups depending on their maturation stage ( Type I: lamellipodia; Type II: immature neuron , sprouting of several minor neurites; Type III: axon and dendrite formation , neuronal polarisation and branching; Type IV: neuron with adult-like morphology , ongoing maturation of differentiated processes ) . For the axonal growth analysis , images of cultured cortical neurons ( DIV2 to DIV6 ) were captured using an Olympus IX70 inverted microscope with a 20x objective and traced manually using Neurolucida software ( MBF Bioscience ) . The length of the axonal tree , length of the longest axon , and number of axon tips were measured using Neurolucida Explorer . For synapse detection , neurons were transfected at DIV2 with GFP and shRNAs . Neurons were fixed and permeabilized at DIV4 , DIV6 , or DIV18 using the Image-iT Fixation/Permeabilization Kit ( Life Technologies ) according to the manufacturer’s protocol . Fixed neurons were then incubated with the primary antibody mouse anti-PSD95 ( 1:1000 , ab2723 , Abcam ) in the presence of 3% normal goat serum at 4°C for 24 h . After washing , the cells were further incubated with the secondary antibody goat anti-mouse DyLight 594 ( 1:1000 , ab96873 , Abcam ) at room temperature for 1 h and visualised using LSM510 confocal microscope ( 63×/1 . 3 water immersion objective ) . The immunofluorescent puncta close to soma colocalizing with GFP were quantified manually . The PSD-95 positive puncta co-localizing with GFP marked neurites were quantified manually . The majority of mice were generated from Wfs+/- male and female [40] breeding pairs . Additional Wfs+/+ mice ( 2 ) were generated from a separate breeding pair on a similar background . Mice were housed in a temperature- and humidity-controlled room . Food and water were available ad libitum , and mice were kept on a 12:12 h light:dark cycle . At 1 y of age , mice were deeply anesthetized and perfused with 0 . 1 M PBS followed by 4% paraformaldehyde ( 4°C ) . Brains were left in skulls to preserve anatomy and incubated in 4% PFA overnight at 4°C and then in PBS until 2 days prior to imaging . Skulls were then placed in 2 mM gadovist in PBS and incubated at 4°C with rocking until imaging . A T2 RARE sequence was used for imaging using a 94/20 Bruker BioSpec small animal MRI with the following parameters: Tr , 900 ms; TE , 47 . 13 ms; imaging matrix , 512 x 360 x 80; spatial resolution , 0 . 0444 x 0 . 03 x 0 . 2 mm for an imaging time of approximately 3 h and 4 min . Volumes were segmented manually by an observer blinded to genotype using ITK-SNAP ( V3 . 2 . 0 ) . For the cortex at the level of the striatum , the volume of cortex from bregma +1 . 70 mm to -2 . 18 mm was quantified . For the optic nerve , volumes were calculated for optic nerve rostral to the optic chiasm . For brain stem ( pons and medulla ) , the most rostral portions of the pons and the most caudal portion of the medulla ventral to the cerebellum are not included in the Paxinos atlas; thus , quantification of brain stem began at the most rostral portion of the pons , ventral to the interpeduncular nucleus and dorsal to the mammillary body ( approximately bregma +3 . 62 mm ) , and ended at the termination of the overlying cerebellum ( approximately bregma -8 . 5 mm ) . Total RNA was isolated from age-matched Wfs1-/- and Wfs1+/+ using the Qiagen RNeasy Mini Kit . Conversion to cDNA was performed by reverse transcription using 1 μg of total RNA with SuperScript III RT kit ( Invitrogen ) . Specific primers were designed for amplification of Mfn2 , Mfn1 , Opa1 , Drp1 , Fis1 , Miro2 , and Cox4 genes . qPCR was performed on an ABI PRISM 7900HT Sequence Detection System . Reactions were performed using an ABI SYBRGreen PCR Master Mix , and raw data were analysed using the ΔΔCt method . All Ct values were normalised to the control gene synaptophysin ( SYN ) . Data are presented as the mean ± SEM . The number of replicates for each type of experiments is given in S2 Table . The D’Agostino-Pearson omnibus test was used to test the normality of distribution . t tests , Mann Whitney test , Wilcoxon signed rank test , one-way ANOVAs followed by Bonferroni posthoc test ( selected pairs ) , or Kruskal-Wallis tests followed by the Dunn test were used to compare differences between experimental samples and control groups . Two-way ANOVAs were used to analyse interactions between two treatments . P values of less than 0 . 05 were considered statistically significant . | Wolfram syndrome ( WS ) is a genetic disorder characterized by diabetes insipidus , diabetes mellitus , optic atrophy , deafness , and brain atrophy . Brain abnormalities occur at the earliest stage of clinical symptoms , suggesting that Wolfram syndrome has a pronounced impact on early brain development . The majority of Wolfram syndrome cases are caused by mutations in the gene Wolfram syndrome 1 ( WFS1 ) , which encodes for a protein localized to the endoplasmic reticulum ( ER ) membrane . However , the clinical symptoms of WS resemble mitochondrial disease symptoms , suggesting strong mitochondrial involvement . In this manuscript , we demonstrate that deficiency of the gene WFS1 triggers an ER-stress cascade , which impairs the function of the IP3-receptor calcium channel , leading to altered calcium homeostasis . The latter leads to dysregulation of mitochondrial dynamics , as characterized by augmented mitophagy—a selective degradation of mitochondria—and inhibited mitochondrial trafficking and fusion , which results in lower levels of ATP and , thus , inhibits neuronal development . These results shed new light onto the mechanisms of neuronal abnormalities in Wolfram syndrome and point out potentially new therapeutic targets . Moreover , our results unravel two rather unexpected links that have an impact beyond the relatively rare Wolfram syndrome . First , relatively mild stress of the ER can seriously disturb mitochondrial dynamics , explaining why alterations at the level of the ER could lead to a mitochondrial phenotype . Second , increased levels of mitophagy , leading to excessive and unwanted mitochondrial clearance , are harmful for neurons . Furthermore , since alterations in the gene WFS1 take place in different neurologic and psychiatric disorders , our work may also have broad implications for understanding the role of mitochondrial dynamics in neuropsychiatric diseases . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"medicine",
"and",
"health",
"sciences",
"luciferase",
"enzymes",
"endoplasmic",
"reticulum",
"cell",
"processes",
"enzymology",
"neuroscience",
"physiological",
"processes",
"homeostasis",
"mitochondria",
"molecular",
"biology",
"techniques",
"bioenergetics",
"cellular",
"structures",
"and",
"organelles",
"research",
"and",
"analysis",
"methods",
"developmental",
"neuroscience",
"animal",
"cells",
"proteins",
"hyperexpression",
"techniques",
"oxidoreductases",
"molecular",
"biology",
"molecular",
"biology",
"assays",
"and",
"analysis",
"techniques",
"gene",
"expression",
"and",
"vector",
"techniques",
"biochemistry",
"cellular",
"neuroscience",
"neuronal",
"morphology",
"cell",
"biology",
"secretory",
"pathway",
"physiology",
"neurons",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"energy-producing",
"organelles"
] | 2016 | Role of Mitochondrial Dynamics in Neuronal Development: Mechanism for Wolfram Syndrome |
Mutations in certain subunits of the DNA repair/transcription factor complex TFIIH are linked to the human syndromes xeroderma pigmentosum ( XP ) , Cockayne's syndrome ( CS ) , and trichothiodystrophy ( TTD ) . One of these subunits , p8/TTDA , interacts with p52 and XPD and is important in maintaining TFIIH stability . Drosophila mutants in the p52 ( Dmp52 ) subunit exhibit phenotypic defects similar to those observed in TTD patients with defects in p8/TTDA and XPD , including reduced levels of TFIIH . Here , we demonstrate that several Dmp52 phenotypes , including lethality , developmental defects , and sterility , can be suppressed by p8/TTDA overexpression . TFIIH levels were also recovered in rescued flies . In addition , p8/TTDA overexpression suppressed a lethal allele of the Drosophila XPB homolog . Furthermore , transgenic flies overexpressing p8/TTDA were more resistant to UV irradiation than were wild-type flies , apparently because of enhanced efficiency of cyclobutane-pyrimidine-dimers and 6–4 pyrimidine-pyrimidone photoproducts repair . This study is the first using an intact higher-animal model to show that one subunit mutant can trans-complement another subunit in a multi-subunit complex linked to human diseases .
The integrity of the DNA molecule can be disrupted by chemical and physical factors that cause diverse types of damage . The nucleotide excision repair ( NER ) pathway operates when DNA is damaged by the covalent addition of methyl groups , the formation of cyclobutane-pyrimidine dimers ( CPDs ) , or the crosslinking of bases in opposite strands [1] . In eukaryotes , NER involves at least 35 proteins that participate in damaged-base recognition , oligonucleotide excision , and molecular repair . An important factor in NER is TFIIH , which also participates in basal transcription mediated by RNA polymerases I and II [2] , [3] . TFIIH is a 10-protein complex composed of two subcomplexes . The subunits XPB , XPD , p62 , p52 , p44 , p34 , and p8 come together to form the core subcomplex of TFIIH , which preferentially participates in NER . The subunits cdk7 , cycH , and MAT1 form the cdk-activating kinase subcomplex ( CAK ) , which is involved in controlling the cell cycle [2] . Together , the core and CAK form the 10-protein TFIIH complex that has a fundamental role in RNA polymerase II ( pol II ) transcription [3] . The TFIIH complex possesses several enzymatic activities that contribute to NER , transcription , and cell cycle control: XPB and XPD , which are both ATPases and DNA helicases; cdk7 , which is a kinase; and p44 , which is an ubiquitin ligase [2] , [4] . In humans , mutations in XPB and XPD subunits cause xeroderma pigmentosum ( XP ) , combined Cockayne's syndrome with xeroderma pigmentosum ( CS/XP ) , and trichothiodystrophy ( TTD ) [1] , [5] . XP is primarily related to defects in NER , CS is associated with deficiencies in transcription-coupled repair ( TCR ) , and TTD is linked to reduced transcription and DNA repair deficiencies [6] . XP patients have sunlight hypersensitivity , abnormal skin pigmentation , and a high predisposition for skin cancer . Individuals afflicted with CS have slow postnatal growth and exhibit defects in nervous system development . TTD patients also have nervous system defects , and have brittle hair , ichthyosis , and fragile nails [6] . A particular form of TTD , termed TTD-A , was recently linked to mutations in the p8 subunit , referred here as p8/TTDA . A characteristic of the cells derived from patients with TTD-A , and XPD-linked TTD , is a reduction in basal TFIIH levels [3] . Intriguingly , p8/TTDA seems not to be an essential gene because humans homozygous for a mutation in the start codon that may result in complete loss of the protein or a truncated peptide survive , as do yeast strains containing disruptions of the homologous gene [3] , [7] . The p8/TTDA gene encodes a 72-amino acid protein that is highly conserved in all eukaryotic organisms [3] , [7] . Transfection of wild-type p8/TTDA rescues TFIIH levels and the UV-sensitive phenotype in p8/TTDA and XPD-derived cultured cells , showing that p8/TTDA is essential for maintaining steady-state levels of TFIIH [8] . p8/TTDA interacts with TFIIH p52 [8] , [9] and XPD subunits [8] , and functions primarily in NER . XPB ATPase activity , which is required for NER , is modulated by the interaction of p8/TTDA and p52 [10] . p8/TTDA exists in two different pools , one in the cytoplasm and one in the nucleus . After DNA damage , p8/TTDA forms a more stable association with TFIIH in nuclei [11] . Recent studies have shown that the fruit fly , Drosophila melanogaster , is a useful model organism for the study of several human diseases . In a number of important cases , mutation or overexpression of a disease-related human gene generates an equivalent phenotype in the fly [12] . An important feature of fly models of human diseases is the ability to use such models in genetic screens to identify new mutations or modifications in gene expression that suppress defective phenotypes [13] . Interestingly , flies carrying mutations in the XPB ( haywire ) and p52 ( Dmp52 ) TFIIH subunits of Drosophila exhibit phenotypes that are comparable to those observed in humans [14]–[17] . In addition , the neurological defects , brittle bristles phenotype , UV-irradiation hypersensitivity , and cuticle defects in flies with defects in TFIIH components appear to exhibit similarities to some symptoms of TTD individuals at the molecular level [14]–[16] . These similarities include reduced transcription of specific genes that are normally required at high levels in terminally differentiated cells [15] , [16] , and a reduction in TFIIH levels [16] , [18] . In this work we report that overexpression of the Drosophila homologue of p8/TTDA ( Dmp8/TTDA ) , suppresses lethal mutations in the Dmp52 and haywire genes . Rescued flies suppress developmental defects , including brittle bristles and thin cuticle , and recover basal TFIIH levels . In addition , transgenic flies overexpressing Dmp8/TTDA are more resistant to UV irradiation than are wild-type organisms and are more efficient in the repair of cyclobutane-pyrimidine dimers ( CPDs ) and 6-4 pyrimidine-pyrimidone photoproducts ( 6-4PPs ) . Collectively , our results open the possibility of a therapy based on enhancement of p8/TTDA function in patients afflicted with TFIIH-related syndromes .
The OreR Drosophila strain was used as a control . All mrn and haywire alleles used in this work have been previously characterized [15]–[17] . The parental strain for the mrn alleles has the red and ebony markers and was used in some UV irradiation experiments as control . The complete Drosophila wild-type p8/TTDA DNA sequence was amplified by PCR and cloned into the pCaSperhsp83 vector and sequenced to verify its integrity . Constructs encoding six histidines at the NH-terminus ( H6-Dmp8/TTDA ) or COOH-terminus ( Dmp8/TTDA-H6 ) of Dmp8/TTDA were also cloned into the pCaSperhsp83 vector . Transgenic flies were constructed using a standard microinjection protocol . The location of transgenes on different chromosomes was determined by balancer mapping . Rescue experiments were performed by crossing pCaSper-hsp83-p8/TTDA transgenic fly lines with mrn and hay alleles , as previously described [16] . In brief , balanced transgenic flies expressing p8/TTDA in the X and second chromosomes and an MKRS/TM3 third chromosome were crossed with different mrn and hay alleles balanced with TM6B . The F1 progeny were crossed to generate homozygous mrn or hay and heteroallelic EP3605/mrn flies containing one or two copies of the transgene in either the second or X chromosome . Third instar wild-type , rescued and transgenic larvae were irradiated at different UV-B light dosages ( Joules/m2 ) using a UV Stratalinker 2400 ( Stratagene ) . The larvae were then allowed to develop into adults and the emerged population was counted . Third instar larvae salivary glands from rescued homozygous mrn mutants and heteroallelic combinations of EP3605 and mrn 1 , 3 and 5 alleles were dissected , immunostained and quantified as previously described [16] . Briefly , using confocal microscopy , representative images of immunostained XPD , XPB , TBP and histones in nuclear sections from wild-type and each Dmp52 genotype were obtained . Nuclear areas ( 156 pixels/nucleus ) were analyzed from each genotype using a photon-counting protocol . Fluorescence-intensity distribution frequencies were obtained and represented as a histogram . Relative fluorescence ratios are presented as a bar chart , which shows the average intensity of XPB/TBP , XPB/histones , XPD/TBP and XPD/histones ( ±standard errors ) in the y-axis for each genotype of Dmp52 mutants and rescued organisms ( x-axis ) . Ten micrograms of genomic DNA isolated from third instar larvae was dotted onto a nitrocellulose membrane and probed with an anti-CPD antibody using a standard Southwestern analysis protocol ( Kamiya Biomedical Company , Seattle WA ) . To measure 6-4PPs , we performed ELISA assays using a specific anti-6-4PP antibody following the standard protocol recommended by the supplier ( Kamiya Biomedical Company ) . In general , total protein soluble extracts were prepared from adult flies and standardized . Then the samples were loaded in 10% SDS-PAGE gels and the proteins transferred to nitrocellulose filters . A specific anti CTD-Ser-5-P antibody was used following standard protocols . Typically , 108 cells were infected with combinations of recombinant baculoviruses expressing XPB , XPD , p62 , p52 ( or the various mutant versions ) , and p44 , p34 , p8 , cdk7 , cyclin H and MAT1 as indicated , and collected 48 h after infection . Cells were washed with phosphate-buffered saline , 30% glycerol and disrupted in 10 ml buffer A ( 20 mM Tris-HCl pH 7 . 5 , 150 mM NaCl , 20% glycerol , 0 . 1% Nonidet P40 , 5 mM b-mercaptoethanol ) using a dounce homogenizer . After centrifugation at 14 , 000× g for 30 min at 4°C , clarified lysates were loaded onto a heparin-Ultrogel column ( Sepracor ) pre-equilibrated in buffer A . After extensive washing with buffer A containing 300 mM NaCl , the proteins were eluted with buffer A containing 500 mM NaCl . The eluted fractions were dialyzed for 2 h against 50 mM Tris-HCl pH 7 . 9 , 50 mM KCl , 20% glycerol , 0 . 1 mM EDTA and 0 . 5 mM dithiothreitol , and immunopurified using the 1H5 anti-p44 antibody [19] . Run-off transcription was carried out as previously described [19] . The dual-incision assay was performed according published methods [20] , [21] . Briefly , repair reactions were carried out in buffer containing 45 mM HEPES pH 7 . 8 , 70 mM KCl , 5 mM MgCl2 , 1 mM dithiothreitol , 0 . 3 mM EDTA , 10% glycerol and 2 mM ATP . Each reaction contained 50 ng XPG , 20 ng XPF/ERCC1 , 10 ng XPC-hHR23B , 50 ng RPA , 25 ng XPA and either 1 . 5 µl of HeLa TFIIH ( Hep fr . IV ) or recombinant TFIIH complexes that included a wild-type or mutant p52 subunit . After pre-incubating for 10 min at 30°C , 50 ng of damaged circular template DNA containing a single 1 , 3-intrastrand d ( GpTpG ) cisplatin-DNA cross-link ( Pt-GTG ) was added and reactions were continued for 90 min at 30°C . The reactions were stopped by rapid freezing . After annealing with 9 ng of the complementary oligonucleotide , a mixture of [a-32P] dNTPs ( 3000 mCi/mmol ) was added and residues were incorporated using Sequenase V2 . 1 ( USB ) . The excised , radiolabeled fragments were separated on a 14% urea–polyacrylamide gel and visualized by autoradiography [19] , [21] .
Our group and others have previously demonstrated that homozygous point mutations in the genes marionette ( mrn ) , encoding the Drosophila homolog of p52 ( Dmp52 ) , and haywire , the Drosophila homolog of XPB , are lethal and share chromosomal fragility and defective developmental phenotypes [14]–[16] . In contrast , organisms with heteroallelic Dmp52 combinations between a P-element insertion near the 5′ region of the Dmp52 gene ( EP3605 ) and mrn point mutations develop into adults , but are sterile , have reduced levels of TFIIH , and present with brittle bristles and cuticle deformations typical of TFIIH-deficient flies [15] , [16] . As p8/TTDA overexpression rescues the UV-sensitive phenotype and TFIIH levels in XPD-deficient cultured cells [8] , we tested whether overexpression of Dmp8/TTDA might rescue the larval lethality and adult phenotypic defects observed in Dmp52 mutant flies . To answer this question , we generated transgenic flies that overexpress Dmp8/TTDA under the control of the HSP83 promoter . This promoter is constitutive and drives moderate levels of transgene expression in all tissues . Transgene expression in the transgenic lines , TTDA1 and TTDA5 , was verified by RT-PCR ( Figure S1 ) . Two additional transgenic lines used in this work , TTDA8 and TTDA9 , express hexahistidine ( H6 ) -tagged recombinant Dmp8/TTDA protein with the tag present at either the N-terminus ( H6-Dmp8/TTDA ) or C-terminus ( Dmp8/TTDA-H6 ) , respectively; the proteins can be detected by Western blotting and immunocytochemistry ( Figure S1 ) . In general , transgenic flies with only a single copy of the transgene expressed approximately 10 times more Dmp8/TTDA transcript than did wild-type flies . In fact , the levels of endogenous Dmp8/TTDA mRNA were very low and were more difficult to detect than were other TFIIH transcripts , suggesting that Dmp8/TTDA could be a limiting factor for TFIIH function ( Figure S1B; unpublished observations ) . To obtain homozygous mrn ( mrn/mrn ) mutants and heteroallelic Dmp52 mutants ( EP3605/mrn ) , carrying one or two copies of the Dmp8/TTDA transgene , we crossed different transgenic flies with flies carrying various mutations affecting the Dmp52 gene [ref 16; see also Figure 1A] . Figure 1B shows that a single copy of the Dmp8/TTDA transgene partially rescued the lethal phenotype of mrn3 homozygous mutant flies . Two copies of the transgene , one on each of the X chromosome and chromosome 2 ( indicated in Figure 1B as TTDA1 and TTDA5 ) , increased the number of organisms rescued . The rescue values were low , but significant , because a transgenic line that expresses a double mutant form of Dmp52 ( Dmp52 ( E340K-R344E ) ) was not able to rescue any mutant lines [ref 16; Figure 1B] . We have previously shown that this double mutant form , generated by site-directed mutagenesis , abolishes the incorporation of XPB into the 10-subunit TFIIH complex , thereby dramatically reducing transcription and NER activity [16] . In addition , a line homozygous for the mrn3 allele carrying three copies of the Dmp8/TTDA transgene is viable and fertile . Thus , overexpression of Dmp8/TTDA was able to rescue the milder homozygous Dmp52 lethal allele , mrn3 , indicating that a 10–fold increase in the expression of Dmp8/TTDA is sufficient to achieve a partial rescue of the lethality of the mrn3 allele . However , two other alleles , mrn1 and mrn5 , which are more deleterious than mrn3 [16] , were not rescued ( Figure 1B ) . The mrn3 allele generates a truncated peptide of 255 amino acids that contains the NH2-terminal portion of the protein . Interestingly , a human version of the mrn3 allele co-expressed with the other recombinant human TFIIH subunits in insect cells is assembled into 6- or 9-subunit complexes ( Figure 2A , indicateted as p52/220st ) and allowed the incorporation of the XPB subunit into TFIIH , one of the functions of p52 ( Figure 2A , lanes 2 and 4 ) . However , these complexes lacked DNA repair and transcription activity in in vitro biochemical assays , both in the presence and absence of Dmp8/TTDA ( Figure 2B , lanes 4 and 8; Figure 2C , lanes 6 and 13 ) , suggesting that suppression of the mrn3 allele in the fly by overexpressed Dmp8/TTDA requires a specific in vivo context that is not readily reconstituted in vitro ( see below ) . Because p8/TTDA also interacts with the XPD component of TFIIH [8] , it is possible that Dmp8/TTDA overexpression in vivo may stabilize the partially functional TFIIH complexes containing the Dmp52 truncated peptide through its interaction with XPD . It is also possible that in the in vivo context , a region of the complex present in the truncated Dmp52 protein may still interact with Dmp8/TTDA . An initial in vitro analysis of the contribution of p8/TTDA to TFIIH activity showed that p8/TTDA participates in NER , but not in transcription [8] . However , it is worth noting that in in vitro transcription assays , Dmp8/TTDA stimulated RNA synthesis of wild-type IIH9 or in the presence of II6+CAK ∼2-fold ( Figure 2C , lanes 3 and 10 ) . The difference between the results presented here and previous reports identifying p8/TTDA as a repair-specific TFIIH subunit [8] may be attributable to differences in the TFIIH preparations [8] . Indeed , this result is in agreement with the discovery of TFB5 ( the p8 yeast homologue ) as component of the transcription Pre-Initiation-Complex [7] . In the yeast system , it has also been demonstrated that nuclear extracts made from yeast carrying a deletion of the TFB5 gene were deficient in transcription in vitro , and this mutant is also unable to activate transcription of inducible genes in vivo [7] . Our results suggest that the human p8/TTDA also increases transcription in vitro through an interaction with TFIIH , possibly by stabilizing TFIIH , however , the mechanism by which p8/TTDA enhances transcription in vitro still requires further study . Heterozygous mrn3/+ flies are viable , but are more sensitive to UV irradiation than are wild-type organisms ( 16; Figure S2 ) . Intriguingly , the UV-irradiation sensitivity of rescued homozygous flies ( TTDA/TTDA; mrn3/mrn3 ) was similar to that of heterozygous mrn3/+ organisms , compared to the parental strain with the same genetic markers ( Figure S2 ) , indicating that although lethality was suppressed by Dmp8/TTDA overexpression , the NER defects of these flies were not completely restored to homozygous wild-type levels . Therefore , mrn3/mrn3 homozygous mutant flies rescued by Dmp8/TTDA overexpression behaved like organisms that contain a single wild-type copy of Dmp52 , and thus NER is only partially recovered . We also found that the Dmp8/TTDA transgenes were able to rescue the sterility phenotype of heteroallelic Dmp52 flies , but the mutant Dmp52 transgene was not ( Figure 1C ) . In all of the heteroallelic combinations tested ( EP3605/mrn1 , mrn3 , or mrn5 ) , sterility was suppressed to different degrees by the Dmp8/TTDA transgene ( Figure 1C ) . In addition , the brittle bristle and cuticle deformation phenotypes commonly observed in the mutated heteroallelic EP3605/mrn flies were also suppressed . Figure 1D shows that EP3605/mrn5 heteroallelic flies have defective bristles and a thin thorax and abdomen . These phenotypes were expressed in nearly 100% of these heteroallelic ( EP3605/mrn1 , mrn3 , or mrn5 ) flies ( Figure 1D; Table 1 ) [16] . However , the presence of a single extra copy of the Dmp8/TTDA transgene ( denoted as EP3605/mrn5; DmTTDA in Figure 1D ) completely suppressed these phenotypes in all flies analyzed ( Figure 1D; Table 1 ) . We have previously demonstrated that these two phenotypes are caused by deficient transcription during fly development and appear to be counterparts of the brittle hair and ichthyosis defects observed in TTD patients , which are also caused by transcriptional deficiencies in TFIIH [15] . Collectively , these results demonstrate that overexpression of Dmp8/TTDA rescues these developmental defects in a complex organism . In addition to testing the ability of Dmp8/TTDA to suppress Dmp52 mutants , we also determined whether overexpression of Dmp8/TTDA might be able to suppress the homozygous lethal phenotype associated with alleles of haywire ( hay ) , which encodes for the XPB-homologous gene . To address this , we used the conditional haync2 ( R652C ) and haync2rv8 ( R652C/E278G ) alleles , which are lethal at 25°C , and the haync2rv7 ( W441stop ) lethal allele [14] , [15] , [17] ( Figure 3A ) . Genetic crosses were performed to obtain flies homozygous for the haync2 , haync2rv8 , and haync2rv7alleles [14] , [15] , [17] , and which contained one extra copy of Dmp8/TTDA and were capable of growth at 25°C . We found that , in this genotypic context , Dmp8/TTDA overexpression rescued viability in the haync2 flies , but not in haync2rv8 or haync2rv7 flies ( Figure 3B ) . The haync2rv8 allele encodes a Hay protein containing two point mutations , and the haync2rv7 mutant generates a truncated Hay protein ( Figure 3A ) ; both alleles are more deleterious than is the haync2 allele [14] , [15] , [17] . Interestingly , an equivalent haync2 mutation introduced into the human XPB , has reduced transcription and repair activities and its interaction with p52 is weakened ( our own unpublished results ) . These results indicate that even though p8/TTDA does not interact directly with XPB , p8/TTDA overexpression can suppress milder mutations in the DmXPB homologue in Drosophila , probably by stabilizing the interaction between XPB and p52 . All thogheter these results are in agreement with a previous report that p8/TTDA overexpression can restore TFIIH levels in TTD-XPD human cells cultured in vitro [8] , and suggest that p8/TTDA overexpression can suppress mutations in different TFIIH subunits . The EP3605 allele was generated by a transposable EP insertion in the non-coding 5′ region of the Dmp52 gene [16] . EP3605 is a hypomorphic allele that may generate low levels of functional Dmp52 [16] . Accordingly , the ability of Dmp8/TTDA overexpression to rescue lethality , fertility defects , brittle bristles , and cuticle deformations in heteroallelic combinations and homozygous mrn3 flies might be explained in terms of an increase in TFIIH basal levels , which are normally low in these mutants . To test this hypothesis , we performed immunofluorescence experiments using antibodies against the XPD and XPB TFIIH subunits in salivary gland nuclei of wild-type Dmp52 heteroallelic mutants and Dmp52 homozygous mutants carrying the Dmp8/TTDA transgenes . TBP and histone H3 antibodies served as internal controls . As predicted , basal XPB and XPD levels were reduced in the Dmp52 mutant cells ( Figure 4 , denoted as mrn3/EP3605 in all panels ) ; however , when Dmp8/TTDA was overexpressed in the rescued mrn3/mrn3 homozygous line , XPD and XPB levels were restored ( Figure 4 , denoted as TTDA/TTDA; mrn3/mrn3 ) . The recovery of basal XPB levels in the rescued flies is of particular importance because p52 is required for the correct assembly of XPB into TFIIH [10] , [16] , [19] , [22] , indicating the presence of more stable TFIIH complexes . These results suggest that Dmp8/TTDA overexpression increases the stability of TFIIH in Dmp52 mutants sufficiently to allow adequate TFIIH function throughout fly development . This is of particular interest as cells derived from patients with defects in p8/TTDA and some with XPD mutations also have low basal TFIIH levels . In addition , it has been reported that the main molecular function of p8/TTDA is to control the steady state levels of TFIIH [8] , [11] . Our data support this hypothesis and suggest that this is the main mechanism that allows the rescue of mutations in other TFIIH subunits by the overexpression of p8/TTDA . Enhanced stability of the TFIIH complex may also lead to an increase in TFIIH DNA repair activities . To test this hypothesis , we exposed wild-type and Dmp8/TTDA-overexpressing transgenic fly lines to different doses of UV radiation . We used three transgenic lines that have a single extra copy of Dmp8/TTDA ( TTDA1 , TTDA8 , and TTDA9 ) , and one line that contains three copies ( TTDA5-3 ) . As shown in Figure 5A , two lines with one copy of Dmp8/TTDA ( lines TTDA1and TTDA9 ) and the line containing three copies were significantly more resistant to UV irradiation than were wild-type flies or the TTDA8 line . Although lethality was high in both transgenic and wild-type flies , survival was 3- to 4–fold higher in transgenic flies; differences were even more dramatic at higher doses ( 175 and 200 J/m2 ) ( Figure 5A ) . These responses were very reproducible and a statistical analysis showed that the differences between wild-type and transgenic flies were significant ( see legend to Figure 5A ) . Interestingly , UV resistance dropped from approximately 65% survival following irradiation at 150 J/m2 to 5% at 175 UV J/m2 ( Figure 5A and Figure S2 ) . This phenomenon was reproducible in different Drosophila strains [16; this work] . It is possible that the DNA damage produced at 175 J/m2 is beyond a critical repair-capacity threshold limit , thus activating check-point systems that prevent the organism ( in this case third instar larvae ) from continuing the developmental process . The increase in UV-irradiation resistance suggests that overexpression of the TFIIH subunit , Dmp8/TTDA , increases NER efficiency in vivo . To confirm this , we analyzed the rate at which CPDs and 6-4PPs were repaired in wild-type flies and in transgenic lines that overexpressed Dmp8/TTDA . Third instar larvae were irradiated at 200 J/m2 and , at different times , CPDs and 6-4PPs in total purified DNA were quantified by Southwestern dot-blot analysis ( CPDs ) and ELISA assays ( 6-4PP ) using antibodies specific to CPDs and 6-4PPs . After UV irradiation for 10 min , the levels of CPDs were similar in the wild-type and transgenic flies ( representative dot blot and summary data obtained from quantification of three independent assays are shown in Fiure 5B ) . However , 4 h after irradiation , the amount of CPDs was significant lower in flies that overexpressed Dmp8/TTDA compared to wild-type flies , a difference that was maintained over time ( Figure 5B ) . We also observed that 6-4PPs were removed more rapidly in the transgenic flies ( Figure 5C ) . In this case , we used ELISA assays instead of Southwestern blots because the results were more reproducible . In Figure 5C , we show an average obtained from three independent measures using the same DNA used for CPDs analysis . A dramatic difference in 6-4PP removal between the transgenic and wild-type flies is evident ( Figure 5C ) . Thus , Dmp8/TTDA transgenic flies removed CPDs and 6-4PPs faster than did wild-type flies , and this increased repair rate correlated with increased resistance to UV irradiation . It is worth noting that the rate of CPD removal in transgenic flies measured here is faster than that seen in some reports using mammalian cells [23] , [24] , [25] . However , another study using our technique in mammalian cells demonstrated a significant removal of CPDs 3 h after UV irradiation of wild-type cells [26] , a time course that is similar to that reported here . It is also important to take into account the fact that most mammalian cell studies have employed cultured fibroblasts to measure CPD removal . To the best of our knowledge , ours is the first study to apply this technique to third instar Drosophila larvae , which are different from in vitro-cultured cells in many respects , one of which is the presence of a thick cuticle that protects the organism and necessitates the use of higher UV doses to produce damage . Another important point to consider is that at the moment of irradiation , these larvae are not only growing but are also preparing to undergo metamorphosis; it possible that the rate of DNA repair could be different during other developmental stages . Intriguingly , the removal of 6-4PPs was very slow in wild-type organisms , only after 16 hrs after irradiation removal was ovserved , however it was very fast in flies overexpressing Dmp8/TTDA . This result is different from a previous report on the rate of 6-4PP repair in cultured Drosophila cells [27] . However , the methods used to measure 6-4PPs were different; in our case , we used the entire organism instead of in vitro-cultured cells and found a similar rate of removal in three independent DNA preparations . The role of TFIIH in NER is to open the double-stranded DNA at the site of damage , a function that depends on the 5′-3′ helicase and ATPase activities of the XPD and XPB subunits , respectively [10] . p52 is required to incorporate XPB into the core of TFIIH and , together with p8/TTDA , may positively regulate XPB-ATPase activity [10] . TFIIH is also important in the recruitment and stabilization of several components of the NER machinery at the DNA-damaged site [22] . In addition , a stable TFIIH complex is an important prerequisite for interaction with specific factors in NER and transcription [28] , [29] . It is possible that p8/TTDA overexpression may help to stabilize TFIIH , thereby enhancing some of TFIIH NER functions . In agreement with this hypothesis , an increase in p8/TTDA levels has been shown to enhance TFIIH repair activity in vitro [8] . Considering the in vitro transcription experiments presented in Figure 2 in light of the fact that overexpression of Dmp8/TTDA enhanced NER in Drosophila , we investigated whether overexpression of Dmp8/TTDA affected transcription in the transgenic flies . To address this question , we measured the levels of Ser-5 phosphorylation in the large subunit of the RNA pol II C-terminal domain ( CTD ) . As has been previously established , the phosphorylation of Ser-5 in the RNA pol II large subunit CTD is a direct measure of the transcriptional activity of TFIIH [30] . To measure Ser-5 phosphorylation levels , we used an antibody that specifically recognizes this modification and protein extracts from wild-type , Dmp8/TTDA-transgenic flies , and rescued flies . Interestingly , CTD Ser-5 phosphorylation levels in the rescued flies are at similar levels as in wild type organisms , when the ratio between the β-tubulin levels and CTD-Ser-5 phosphorylated is compared in each phenotype ( bottom of Figure 5D ) . However , there was no significant increase in the phosphorylation of the CTD in transgenic flies with a wild-type background ( Figure 5D ) . These results suggest that Dmp8/TTDA is not a limiting factor in transcription . There have been reports that a mutant in one subunit of a multifunctional complex can be trans-complemented by overexpression of another subunit , but only in cultured cells or in unicellular organisms [8] , [31] , [32] . In this work , we show that overexpression of Dmp8/TTDA can suppress a mutation in other TFIIH subunits and enhance UV-irradiation resistance in a living multicellular organism . Notably , developmental defects that appear in mutant adult organisms with defects in Dmp52 were suppressed . Some of these Dmp52 mutant phenotypes , such as cuticle deformations and brittle bristles , are caused by transcriptional defects during fly development and are , in many ways , quite homologous to some TTD manifestations [15] , [16] . An enabling observation and important motivation for this work is evidence that overexpression of the human p8/TTDA gene in human fibroblasts derived from patients with TTD caused by a mutation in the XPD gene ( mutant: XPDR112H/R112H ) can suppress some of the phenotypes observed in this cell line [8] . However , because of the limitations of in vitro-cultured cell systems , defects that can be generated during development were not studied; the only phenotypes that could be analyzed were TFIIH levels and UV-irradiation sensitivity . Results obtained in cultured cells cannot always be extrapolated to a complete animal . TFIIH participates in three important and highly regulated functions during animal development that must be coordinated with differentiation programs at different developmental times . Mutations in TFIIH that reduce TFIIH functions do not necessarily have the same effects in different cell types . This is observed in humans , where some tissues or developmental processes ( e . g . , neurological defects ) are preferentially affected depending on which subunit is mutated and where in the protein the amino acid change occurs [5] , [6] , [15] , [16] . In this context , a recent work reported minimal differences in gene expression in proliferating fibroblasts from TTD , XPD , and normal donors , indicating that cultured cells do not recapitulate all the differences found in patients afflicted with different TFIIH-related syndromes [33] . Many of the phenotypes observed in flies with different Dmp52 and hay alleles arise because of defects that accumulate during development . These defects can be partially corrected by the overexpression of Dmp8/TTDA; in other words , the suppression of Dmp52 and hay mutations by Dmp8/TTDA is sufficient to allow the developmental program in a complex animal to run to completion . The results presented here open the possibility that new treatments geared toward enhancing p8/TTDA function might stabilize TFIIH in patients with deficiencies of this DNA repair/transcription complex . This might be accomplished through the design of new drugs that enhance p8/TTDA function or by gene therapy strategies based on p8/TTDA overexpression . The effectiveness of either strategy may ultimately depend on resolving the three-dimensional structure of the interacting surfaces of p8/TTDA and other TFIIH subunits [34] . Mouse models , such as transgenic mice that carry XPD alleles known to cause TTD or XP/CS in humans , and which manifest some of the typical TTD or XP/CS phenotypes [35] , [36] , provide additional tools , making it possible to determine if the overexpression of p8/TTDA is able to rescue specific TFIIH-defective phenotypes . Many cellular functions , including transcriptional activation , chromatin remodeling , and histone modification are mediated by multi-subunit protein complexes . Some of the subunits in these complexes are relatively small and have no known function , although mutations in some of these complex components have been linked to human diseases . It has been suggested that p8/TTDA may act as a kind of small chaperone protein to stabilize TFIIH [11] , raising the possibility that , like p8/TTDA , some of these uncharacterized complex components may serve to maintain the stability and the steady state levels of the corresponding complexes . Thus , it will be important to determine whether proteins in other multi-subunit complexes possess p8/TTDA-like functions . | TFIIH participates in RNA polymerase II transcription , nucleotide excision repair , and control of the cell cycle . In humans , certain mutations in the XPB and XPD subunits of TFIIH generate the syndromes trichothiodystrophy ( TTD ) , xeroderma pigmentosum ( XP ) , and Cockayne's syndrome ( CS ) . In contrast , mutations in the p8/TTDA subunit have been linked only to TTD . Cells derived from TTD patients with defects in p8/TTDA have reduced levels of TFIIH . Therefore , it has been proposed that the main function of p8/TTDA is to stabilize and maintain steady-state levels of TFIIH . In Drosophila , mutations in Dmp52 and haywire genes generate phenotypes that share similarities with those associated with mutations in their human counterparts , including reduced TFIIH levels . We report that p8/TTDA overexpression suppressed accumulated developmental defects associated with mutations in the Dmp52 and haywire genes . We also provide evidence suggesting that the rescue of these defects is , in part , because of the recovery of normal TFIIH levels in mutant flies . These results indicate that overexpression of p8/TTDA trans-complemented mutations in other TFIIH subunits and suppressed defects accumulated during fly development . The overexpression of p8/TTDA in wild-type flies increased their UV irradiation resistance , apparently because of more efficient nucleotide excision repair . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results/Discussion"
] | [
"developmental",
"biology",
"molecular",
"biology/transcription",
"initiation",
"and",
"activation",
"cell",
"biology/developmental",
"molecular",
"mechanisms",
"genetics",
"and",
"genomics/disease",
"models",
"molecular",
"biology/dna",
"repair"
] | 2008 | p8/TTDA Overexpression Enhances UV-Irradiation Resistance and Suppresses TFIIH Mutations in a Drosophila Trichothiodystrophy Model |
Identification of the determinants of pathogen reservoir potential is central to understand disease emergence . It has been proposed that host lifespan is one such determinant: short-lived hosts will invest less in costly defenses against pathogens , so that they will be more susceptible to infection , more competent as sources of infection and/or will sustain larger vector populations , thus being effective reservoirs for the infection of long-lived hosts . This hypothesis is sustained by analyses of different hosts of multihost pathogens , but not of different genotypes of the same host species . Here we examined this hypothesis by comparing two genotypes of the plant Arabidopsis thaliana that differ largely both in life-span and in tolerance to its natural pathogen Cucumber mosaic virus ( CMV ) . Experiments with the aphid vector Myzus persicae showed that both genotypes were similarly competent as sources for virus transmission , but the short-lived genotype was more susceptible to infection and was able to sustain larger vector populations . To explore how differences in defense against CMV and its vector relate to reservoir potential , we developed a model that was run for a set of experimentally-determined parameters , and for a realistic range of host plant and vector population densities . Model simulations showed that the less efficient defenses of the short-lived genotype resulted in higher reservoir potential , which in heterogeneous host populations may be balanced by the longer infectious period of the long-lived genotype . This balance was modulated by the demography of both host and vector populations , and by the genetic composition of the host population . Thus , within-species genetic diversity for lifespan and defenses against pathogens will result in polymorphisms for pathogen reservoir potential , which will condition within-population infection dynamics . These results are relevant for a better understanding of host-pathogen co-evolution , and of the dynamics of pathogen emergence .
Understanding the complex interplay of factors resulting in pathogen emergence has been a major goal of evolutionary ecology for the last twenty years , as emerging infectious diseases often have a high impact in human and animal health , agriculture and conservation [1]–[5] . Generally , emergent pathogens are multi-host pathogens that spill-over onto a new host population from one or more epidemiologically connected populations in which the pathogen can be permanently maintained , that is , from a reservoir host sensu Haydon et al . [2] . Hence , identifying the causes that determine the reservoir potential of a host , i . e . , its ability to sustain pathogen populations for transmission to the target host , is central for understanding emergence and , more generally , infection dynamics . Emergent pathogens are often vector-transmitted [1] , [3] , [4] , so that parameters associated with the triple interaction host-pathogen-vector should be taken into consideration for predicting host reservoir potential . For vector-transmitted pathogens , three epidemiological parameters have been underscored as modulating host reservoir potential: i ) the probability that a vector acquires the pathogen when feeding on an infected host ( host competence ) , ii ) the probability that a host is infected by a feeding vector that carries the pathogen ( host susceptibility ) , and iii ) the ability of the host to sustain vector populations [6]–[8] . These parameters vary among host species and genotypes [2] , [6]–[9] , and knowing which factors determine such variation will facilitate identifying the potential of a host as a reservoir for pathogen emergence . Host lifespan has been identified as a trait related to host reservoir potential . The rationale for linking host lifespan and reservoir potential is that evolution of defenses against pathogens has a fitness cost for the host in terms of other advantageous life history traits , such as fecundity or survival [10]–[13] . Because the probability of becoming infected is higher in long-lived host individuals than in short-lived ones , disease prevalence will be higher in populations of long-lived hosts . Thus , long-lived hosts will be under higher selection pressures for developing costly defenses than short-lived hosts . Model analyses under different scenarios do indeed show that long-lived hosts generally will invest more in defense against pathogens [14] . Major forms of defense are resistance , which has been defined as mechanisms that reduce infection and/or pathogen multiplication in the infected host [15] , [16] , and tolerance , defined as mechanisms that reduce the negative impact of infection on host fitness , i . e . , that reduce pathogen virulence [15] , [17] . By the same rationale , longer-lived hosts would be expected to have developed defenses that would reduce their ability to sustain vector populations . However , the relationship between lifespan and defense is a complex one and , because the demographic turnover of short-lived hosts is high and that of long-lived ones is low , it may be influenced by demographic factors , notably by population density and composition [14] , [18] . This is in agreement with experimental data showing a link between host density and defense evolution [19]–[21] . Despite the attention received from theoreticians , experimental analyses of the relationship between host lifespan and defense are scarce . For plants , there is evidence that short-lived species of grasses ( Poaceae ) have a high reservoir potential of a generalist plant virus for long-lived hosts [22] , [23] . Also , it has been experimentally shown that defenses reducing susceptibility to infection by that virus was lower in short-lived ( annual ) that in long-lived ( perennial ) species of grasses , while competence was not explained by lifespan [8] , [24] . To our knowledge , the relationship between host lifespan and reservoir potential has not been analyzed at the within-species diversity level . Here we address the hypothesis that there is a trade-off between defense to a vectored pathogen , and to its vector , and host lifespan , which would explain pathogen reservoir potential . Specifically we ask the following set of questions: i ) is host lifespan related to competence , susceptibility and the ability to sustain vector populations ? ii ) is host lifespan related to reservoir potential ? , and iii ) does reservoir potential depend on the density and genetic composition of the host population ? . For this we used an experimental approach for estimating realistic values of the parameters determining host reservoir potential , as a basis for model analyses of the factors that modulate reservoir potential . We focused our research on a plant-virus system: the wild plant Arabidopsis thaliana L . Heynh . ( Brassicaceae ) , its natural pathogen Cucumber mosaic virus ( Bromoviridae ) , and the virus insect vector Myzus persicae Sulzer ( Aphididae ) , an interactive assembly of biological components found in nature . A . thaliana ( from here on , Arabidopsis ) has been for a long time the model organism of choice for plant molecular genetics , and is increasingly used in analyses of plant ecology , including the evolutionary ecology of plant-pathogen interactions [e . g . 25–29] . Arabidopsis is an annual species , presently distributed worldwide after experiencing an expansion from its native range in Eurasia and North Africa [30] . The genetic structure of Arabidopsis in the Iberian Peninsula has been studied in detail , demonstrating that it is a centre of genetic diversity for this species [31]–[33] . Demographical analyses carried on in the Iberian Peninsula indicated that Arabidopsis plants flower and complete their life cycle in spring , and that populations are built of two or one cohorts of plants that either germinate in the autumn and overwinter as rosettes , or germinate in the spring [29] , [34] . Also , populations are genetically heterogeneous and are composed of short-lived genotypes , which do not require vernalization to complete their life cycle , and of long-lived ones , that usually require vernalization and belong to the autumn cohort [32] , [35] . Cucumber mosaic virus ( CMV ) is a single-stranded , messenger sense RNA virus , with a three-partite genome encapsidated in isometric particles . CMV has an extremely broad host range , infecting over 1200 species in more than 100 plant families . CMV is efficiently transmitted by more than 75 species of aphids ( Hemiptera: Aphididae ) [36] , in a non-persistent , stylet-borne manner , i . e . , CMV does not infect the insect vector , instead , particles are retained in the maxillary stylets , allowing the aphid to transmit the virus for a short time ( less than 6 h ) after acquisition . The green peach aphid ( M . persicae ) , a cosmopolitan aphid species , is an important component of the aphid populations feeding on plants in a variety of habitats in Spain ( http://www . cabi . org/isc/datasheet/35642 ) , including those in which wild Arabidopsis populations are present . M . persicae is one of the most efficient vectors for CMV , and is frequently used in transmission experimentation [37] . Analyses of virus infection in six wild Arabidopsis populations from different habitats of Central Spain have shown that CMV was the major viral pathogen , with prevalence reaching over 70% , according to the population and the year [29 and unpublished data] . Importantly , while maximum virus prevalence occurred at early developmental stages of the plants , plants remained infected and infectious for the rest of their life . CMV is also transmitted through the seed , with efficiency varying largely according to the plant species and genotype , and it has been reported in Arabidopsis to vary between 2 to 8% [38] . It has been shown that Arabidopsis genotypes differ in their ability to sustain populations of M . persicae [39] . Also , Arabidopsis genotypes differ in resistance and tolerance to CMV , and previous work in our group has shown that tolerance to CMV infection was correlated positively with lifespan in Arabidopsis [40] , [41] . We used an experimental approach comparing two Arabidopsis genotypes that differ largely in lifespan , for the following traits: competence as a source for infection , susceptibility to infection , ability to sustain vector populations and rates of seed transmission . These data were then used as a basis to estimate realistic values of the parameters for the analysis of host reservoir potential using a model in which the dynamics of infected plants and of viruliferous vectors were jointly considered . Our results show that the short-lived and the long-lived genotypes differed in susceptibility and in their ability to sustain vector populations , which resulted in a higher reservoir potential for the short-lived host . However , reservoir efficiency of the short-lived genotype was modulated by the density of host plant and vector populations , and by the genetic composition of the host population .
The CMV strain LS-CMV used in this work was derived from biologically active cDNA clones [42] ( Accession numbers AF416899 , AF416900 and AF127976 for genomic RNAs 1 , 2 and 3 , respectively ) . In vitro transcripts were multiplied in Nicotiana clevelandii plants , virions were purified as in Lot et al . [43] and viral RNA was extracted by virion disruption with phenol and sodium dodecyl sulphate . Two genotypes of Arabidopsis , Landsberg erecta ( Ler ) and Llagostera ( Ll-0 ) , were chosen for the very different span of their life-cycle and tolerance to LS-CMV infection [40] , [41] . The genotype Columbia glabrata 1 ( Colgb1 ) was used as an additional virus source for aphid transmission of CMV to Ler and Ll-0 . These three Arabidopsis genotypes were initially multiplied simultaneously under the same greenhouse conditions to minimize maternal effects . For experiments , Arabidopsis seeds were surface-sterilized , plated on one-half-strength Murashige and Skoog basal salt mix medium [44] , 1% ( w/v ) sucrose , 0 . 8% ( w/v ) Bacto agar , and stratified in the dark at 4°C for 96 h . Plates were then transferred to a growth chamber at 22°C under long day conditions ( 16 h light/8 h dark ) and 65–70% relative humidity . Five days-old seedlings were transplanted into 96 well trays with a mix 3∶1 , peat∶vermiculite , and after 10 days were transferred into individual 10 cm diameter pots containing the same substrate , in order to minimize spatial and resource limitation . Plants were grown in a growth chamber at 22°C under normal light ( 220–250 µmol . S−1 . m−2 ) and long day conditions with 65–70% relative humidity . Generally 2–3 days post-transfer , when plants had 4–5 leaves ( stages 1 . 04–1 . 05 as in Boyes et al . [45] ) , three rosette leaves per plant were mechanically inoculated with a total of 15 µl of 100 µg/ml suspension of LS-CMV RNA in 0 . 1M Na2HPO4 . Only 15 µl of 0 . 1M Na2HPO4 were applied to mock-inoculated controls . Ten days post-inoculation ( dpi ) , three circles of 4 mm diameter were cut from three randomly chosen systemically infected leaves . In this sample , systemic infection was confirmed by ELISA with PathoScreen CMV Kit ( Agdia , Elkhart , IN , USA ) , and virus accumulation was quantified by quantitative PCR ( qRT-PCR ) . For this , total RNA extracts were obtained using TRIzol reagent according to manufacturer's protocol ( Life Technologies , Carlsbad , CA , USA ) , and then utilized with Brilliant III Ultra-Fast SYBR Green QRT-PCR Master Mix following manufacturer's recommendations ( Agilent Technologies , Santa Clara , CA , USA ) in a final volume of 10 µl . Assays were performed in triplicate on a LightCycler 480 II real-time PCR system ( Roche , Indianapolis , IN , USA ) . Primers CMV-CP LS Q Fwd ( TAAGAAGCTTGTTTCGCGCATTC ) and CMV-CP LS Q Rev ( CGGAAAGATCGGATGATGAAGG ) were designed to amplify 106 nt of the LS-CMV coat protein ( CP ) gene ( GenBank: AF_127976 ) and primers UBI/PE4Fwd ( AATGCTTGGAGTCCTGCTTG ) and UBI/PE4 Rev ( CTTAGAAGATTCCCTGAGTCGC ) amplified 107 nt of the peroxin4 mRNA ( GenBank: NM_122477 ) used as the loading internal control . Quantification was expressed as pg of viral RNA per ng of total RNA . No-template reactions were included in each trial . Thermal parameter for RT-PCR amplification were 50°C for 10 min , 95°C for 3 min and 40 cycles of 95°C for 5 s and 60°C for 10 s . Dissociation curves were generated to ascertain that only one single product was produced and detected in each case . Time to bolting and to flowering was estimated as the number of days from the end of stratification until the appearance of the reproductive meristem and the first open flower , respectively . Rosette longevity was estimated as the number of days from the end of stratification until 50% of rosette leaf-senescence . Total plant lifespan was estimated as the time in days from the end of stratification until 50% of siliques had shattered . To quantify leaf mass per unit area ( LMA , g×m−2 ) , leaves were harvested at flowering and digitally captured for area measurement using the software ImageJ 1 . 47v ( NIH , USA , http://rsbweb . nih . gov/ij/index . html ) . Then , leaf dry weight was determined after maintaining the leaves at 60°C until constant weight was reached . Rosette relative growth rate ( cm/ ( day×cm ) ) was estimated by digital capture and further measure of rosette diameter increase every 2 days for a period of 12 days post inoculation . LMA and relative growth rate are parameters that relate to quick-return and slow-return physiological phenotypes , which may influence host competence , susceptibility , and the ability to support vectors [8] , [24] . To assay germination potential , the same day as siliques were ripe , seeds were collected and plated onto germination media without surface-sterilization . Seeds were then either directly transferred to a light/dark regime as previously mentioned or they were submitted to stratification prior to the light/dark protocol . The percentage of germinated seeds was determined 4 days post-transfer to the light/dark cycle . To analyze the effect of virus infection on seed germination , a period of 10 months dormancy ( storage in the dark at room temperature ) was applied to the seeds before carrying out the regular aforementioned surface-sterilization/stratification protocol . Seed transmission of LS-CMV was estimated by qRT-PCR , testing fifteen biological replicates , consisting of a mix of six 5-day-old seedlings . Virus transmission rate for a single seed was then estimated using the expression reported by Gibbs and Gower [46] , , where p is the probability of virus transmission by a single seed , y is the number of positive samples , n is the total number of samples assayed ( n = 15 ) , and k is the number of seedlings per sample ( k = 6 ) . Experiments were performed using a clonal population of the CMV aphid vector M . persicae derived from a single virginiparous apterous female collected in a pepper crop at Alcalá de Henares ( Madrid , Spain ) in 1989 . In order to obtain apterous or alate adult aphids adapted to Arabidopsis , two aphid colonies were maintained at ICA-CSIC ( Madrid , Spain ) ( Lat . 40°43′97″N , Long . 3°68′69″W , Alt . 710 m ) on an equal mixture of Ler and Ll-0 plants in rearing cages in environmental growth chambers under different conditions: i ) For producing apterous aphids , the colony was maintained at low aphid density at 23/18°C ( light/dark ) temperature , 14 h/10 h ( light/dark ) photoperiod and 60–80% relative humidity , and ii ) Alate aphids were reared at 20/16°C ( light/dark ) temperature , 12 h/12 h ( light/dark ) photoperiod and 60–80% relative humidity . Newly emerged alates of 0–48 h of age were used for experiments . Host plant preference assays were performed with non-viruliferous alate aphids in several dual-choice assays within a 1 m3 arena . Aphids were offered as choices either mock inoculated Ler and Ll-0 plants , or LS-CMV-infected Ler and Ll-0 plants , or mock-inoculated and LS-CMV infected plants of either genotype . Twenty plants , ten of each choice class , were randomly placed in each arena during the trial . Two hundred winged aphids were placed in a flight platform similar to the one described by Fereres et al . [47] and released 0 . 5 m above the test plants . Settled alate individuals , as well as nymphs produced on each plant , were counted 24 h after their release . To estimate the capacity of mock-inoculated and LS-CMV-infected Ler and Ll-0 plants to sustain vector populations , the intrinsic rate of natural increase of aphids was determined . Apterous adults of the M . persicae colony of equal age and weight were used . For synchronization , three apterous adult females were placed on each Arabidopsis test plant and confined using a plastic cup . The following day , adults were removed and three newly born nymphs were kept on each plant , which reached adulthood 7–8 days later . At that point , only one adult was kept per test plant , and every 24 h the newly born nymphs were counted and removed to avoid crowding effects that could influence the reproductive potential . The intrinsic rate of natural increase of the aphid population ( rm ) was calculated according to the equation proposed by Wyatt and White [48] , , where Nd is the number of progeny produced by an adult during d days , d being the pre-reproductive period ( number of days from birth to first reproduction ) . Between 25 and 34 plants per treatment , for a total of 116 plants , were used in this study . Transmission assays were performed as described by Fereres et al . [49] . Briefly , groups of 30–40 apterous individuals of M . persicae were collected and after a pre-acquisition starvation period of one hour were released on LS-CMV infected source plants . All infected plants used as sources for transmission exhibited a saturated optical density ( OD ) in an ELISA test for CMV . Following a 10 min acquisition access period , groups of 5 aphids were transferred to each of the 3 week-old Arabidopsis target plants . A 24 h inoculation access period was allowed before spraying plants with imidacloprid ( Confidor , Bayer ) . Plants were then transferred to an aphid-free growth chamber at 22°C under long day cycle . Three weeks post-inoculation , plants were tested by ELISA for LS-CMV infection . Samples were considered positive when their OD was 3 times higher than the negative control's OD after 24 h of incubation . A total of 37 source plants were sequentially tested , with transmission assayed to 252 Ler and 252 Ll-0 target plants . Transmission rate by a single aphid vector was estimated based on the Gibbs and Gower equation [46] ( see above ) , from the number of positive samples ( y ) , the total number of samples assayed ( n ) , and the number of aphids used per transmission trial ( k = 5 ) . All the above experiments involving M . persicae were performed twice , and results are presented as the mean values of both experiments . Statistical analyses were performed using the statistical software package Statgraphics Centurion version 15 . 1 . 02 ( StatPoint technologies , Inc . , Warrenton , VA ) . The data sets were analyzed using analysis of variance ( ANOVA ) and transformation by was applied when necessary for data normalization .
The two genotypes of Arabidopsis used in this work , Ler and Ll-0 , presented large differences in their life history . Lifespan ( from germination until senescence , see Material & Methods ) of Ler plants was of 57 . 25±0 . 70 days ( d ) , while it was significantly longer , 87 . 88±1 . 06 d , for Ll-0 ( Table 1 ) ( F1 , 16 = 581 . 15 , P≤10−5 ) . Thus , lifespan defined a short-lived ( Ler ) and a long-lived ( Ll-0 ) genotype . Differences in lifespan between Ler and Ll-0 were also associated with differences in each of any other temporal life-history trait measured ( F1 , 16≥234 . 37 , P≤10−5 ) , such as time to bolting , flowering time or time to first silique shattered ( Table 1 ) . In addition to differing in temporal life-history traits , Ler and Ll-0 plants also differed broadly in morphology . As reported [40] , [41] the allometric relationship between vegetative ( rosette ) and reproductive ( inflorescence ) parts was much larger for Ll-0 , which had large rosette built of more than 150 leaves , as compared with the small rosettes of Ler plants , with less than 10 leaves ( Figure S1 ) . Also growth rate estimated by rosette diameter increase per day was higher in Ll-0 than in Ler , but the leaf mass per unit area did not differ between both genotypes ( Table 1 ) . Ler and Ll-0 plants also differed in seed germination potential . While seed germination rate was similar for Ler and Ll-0 after a ten month dormancy period ( Table 1 ) , seeds of Ler had a germination rate of 95% as soon as siliques reached maturity . This was not the case for Ll-0 , which seeds showed a germination rate of 0% at silique maturation . These data indicate a lack of stratification or dormancy requirements for Ler seeds , and the need of a dormancy period for Ll-0 seed germination . To test if lifespan was related to LS-CMV multiplication in Ler and Ll-0 genotypes , virus accumulation in systemically infected leaves ten dpi was analyzed . Virus accumulation did not differ significantly in Ler and in Ll-0 plants ( 5 . 35±1 . 74 pg/ng and 2 . 57±0 . 28 pg/ng , respectively , F1 , 16 = 2 . 48 , P = 0 . 1376 ) . Next , we tested host susceptibility to horizontal transmission , i . e . , the probability of the host becoming infected by a viruliferous M . persicae vector . Transmission experiments were performed using Ler and Ll-0 plants as targets for transmission , and Ler , Ll-0 and Colgb1 plants as sources for virus acquisition by aphids . Data ( Table 2 ) show that Ler and Ll-0 plants were similarly susceptible to LS-CMV aphid transmission when the aphids acquired the virus in plants of the same genotype ( 9 . 67±2 . 89 and 8 . 72±1 . 91% transmission for Ler and Ll-0 , respectively , F1 , 74 = 0 . 59 , P = 0 . 4431 ) . Nonetheless , susceptibility to transmission in Ll-0 plants was dependent on the inoculum source , being significantly lower ( F2 , 37 = 4 . 35 , P = 0 . 0208 ) when aphids acquired the virus in Ler plants ( 3 . 83±0 . 85% ) than when the virus was acquired in Ll-0 or in Colgb1 plants ( 8 . 72±1 . 91% and 10 . 48±2 . 13% , respectively ) . On the contrary , susceptibility of Ler plants was not affected by the genotype of the plants in which the virus was acquired ( F2 , 37 = 0 . 28 , P = 0 . 7548 ) . Last , susceptibility to vertical transmission through the seed was similar for Ler and Ll-0 plants , the rate of single seed transmission being of 1 . 98±0 . 70 and 2 . 47±0 . 82% , respectively ( values from eight assayed mother plants , F1 , 16 = 0 . 21 , P = 0 . 6537 ) . The effects of LS-CMV infection differed in the two Arabidopsis genotypes analyzed ( Table 1 ) . Infection resulted in an increase of the lifespan of Ll-0 plants to 99 . 25±2 . 19 d , in comparison with mock-inoculated plants ( F1 , 16 = 22 . 07 , P = 0 . 0003 ) , while the lifespan of infected Ler plants did not differ from that of mock-inoculated plants ( 58 . 5±0 . 5 days , F1 , 16 = 2 . 11 , P = 0 . 1685 ) . The longer life-span of infected Ll-0 plants is the result of reduced growth rates . Thus , infection resulted in a reduction of the relative rate of rosette growth for Ll-0 plants ( F1 , 10 = 10 . 53 , P = 0 . 0118 ) but not for Ler plants ( F1 , 10 = 0 . 11 , P = 0 . 7476 ) , and in a delay in the development of the inflorescence , as shown by the time to flowering and to silique maturation ( Table 1 ) . Neither the leaf mass per unit area ( F1 , 6≤0 . 41 , P≥0 . 5587 ) or the seed germination rate after a 10 month dormancy ( F1 , 16≤0 . 16 , P≥0 . 6930 ) were affected by infection in either genotype . Two parameters related to host competence , i . e . , to the host capacity as a source of virus for vector transmission , were also estimated for Ler and Ll-0 plants: their capacity to attract vectors and to sustain vector populations . Results from dual free-choice experiments showed that a similar number of winged morphs of M . persicae had settled on mock-inoculated Ler and Ll-0 plants , and a similar number of nymphs were produced in both genotypes after 24 h , which is an indication of the time an aphid spends on the plant on which it has landed ( Table 3 , Cage 3 , F1 , 40 = 0 . 92 , P = 0 . 3424 and F1 , 40 = 1 . 47 , P = 0 . 2333 respectively ) . When mock-inoculated and infected plants were compared , it was found that more aphids settled on mock-inoculated than on LS-CMV-infected plants of both genotypes ( Table 3 , Cage 1 and Cage 2 ) . While this preferential settlement was only marginally significant for each plant genotype , Ler or Ll-0 , when analyzed separately ( F1 , 40 = 1 . 72 , P = 0 . 1981 and F1 , 40 = 3 . 37 , P = 0 . 0743 for Ler and Ll-0 respectively ) , it was significant when data from both genotypes were pooled together ( F1 , 80 = 4 . 86 , P = 0 . 0304 ) , indicating that virus infection reduced aphid preference after 24 h . Similarly , more nymphs were produced in mock-inoculated plants than in infected ones ( F1 , 40 = 1 . 92 , P = 0 . 1738 and F1 , 40 = 9 . 20 , P = 0 . 0043 for Ler and Ll-0 plants , respectively , the difference being significant when data from both genotypes were pooled , F1 , 80 = 4 . 40 , P = 0 . 0391 ) . Interestingly , more winged adult aphids were recovered from infected Ler than from infected Ll-0 plants ( Table 3 , Cage 4; F1 , 40 = 4 . 53 , P = 0 . 0398 ) , and those winged adults had produced more nymphs after 24 h ( F1 , 40 = 6 . 98 , P = 0 . 0119 ) . We also estimated the intrinsic rate of natural increase ( rm ) , and other reproduction-related parameters , of the M . persicae population on mock-inoculated and LS-CMV-infected plants of Ler and Ll-0 genotypes . Results in Table 4 show that the aphid's pre-reproductive period was affected neither by plant genotype nor by infection status ( F3 , 116 = 0 . 99 , P = 0 . 3996 ) . On the other hand , a trend was observed towards higher numbers of total nymphs produced , daily fecundity and rm from adults feeding on infected plants as compared with those from mock-inoculated plants . Although this trend was not statistically significant for all parameters ( see Table 4 ) , it was consistent for both genotypes ( F1 , 53≥2 . 52 , P≤0 . 1182 and F1 , 63≥2 . 37 , P≤0 . 1290 for Ll-0 and Ler , respectively ) and statistically significant when data from both genotypes were pooled together ( F1 , 116≥5 . 41 , P≤0 . 0217 ) . Interestingly , when plants were considered according to genotype , and not to infection status , it was found that Ler plants were able to better support aphid population growth than Ll-0 plants , as rm values differed significantly ( 0 . 3697 vs . 0 . 3506 , F1 , 116 = 3 . 89 , P = 0 . 0509 ) .
To analyze the differences between short- and long-lived genotypes of Arabidopsis in reservoir potential , and the effects of host plant density and population composition in this trait , we used a model in which the dynamics of the plant and vector populations were jointly considered , based on that proposed by Madden et al . [50] . In this model , the rate of plant infection depends on the interaction of susceptible non-infected plants with infectious vectors , hence on the density of infected plants and infectious vectors . Our model differs from that of Madden et al . [50] in that: i ) We considered the plant population divided into two classes , susceptible non-infected ( S ) , and infected ( I ) . We did not consider a separated class of infected-non-infectious plants ( i . e . we do not consider a latent period ) since the rate of colonization of host tissues and organs by the virus , which determines transmissibility , is not genotype-dependent ( our unpublished results ) . Neither did we consider a class of recovered plants , as CMV causes systemic chronic infections so that plants , once infected , remain so until the end of their life cycle . Also , infected plants were considered to be a source of infection for aphids until the end of their life cycle , because aphids fed on stalks and cauline leaves until senescence ( our unpublished observations ) . ii ) The aphid vector population was divided into two classes , non-viruliferous ( X ) , and viruliferous aphids ( Z ) , i . e . , aphids that had acquired the virus by feeding on infected , I , plants and were able to transmit it by feeding on non-infected , S , plants . Since CMV is transmitted non-persistently , there is no latent period for transmission . iii ) A major difference with Madden et al . [50] model is that virulence , ν , expressed as the effect of infection on plant mortality ) was considered in our model . For a single host , the dynamics of the model is described by the following equations: ( 1 ) ( 2 ) ( 3 ) ( 4 ) These equations represent the variation with time ( days ) of the density of non-infected susceptible plants , S ( Eq . 1 ) , of infected plants , I , ( Eq . 2 ) , both expressed as plants×m−2 , and of virus-free aphids , X , ( Eq . 3 ) and viruliferous aphids , Z , ( Eq . 4 ) , expressed as aphids×plant−1 . K indicates the maximum density of the plant population . Parameter m indicates the per capita plant mortality rate , and parameter v indicates the variation of plant mortality rate due to infection ( i . e . , virulence ) . The transmission rate is decomposed into two parameters , Φ , which indicates the number of susceptible plants visited per day by an aphid , and b , which represents the probability of virus transmission per vector visit . Similarly , parameter a indicates the probability that a vector acquires the virus at each visit . Because transmission is non-persistent , viruliferous vectors lose the virus with a rate τ , returning to the X class of virus-free vectors . Last , α indicates the per capita mortality of aphids , which is not affected by their viruliferous state . This model was extended to two hosts and its dynamics for Host 1 are described by the set of equations: ( 5 ) ( 6 ) ( 7 ) ( 8 ) Subscripts 1 and 2 denote the host . Equations 5–8 describe the variation with time ( days ) of the density of susceptible plants , Si ( Eq . 5 ) , of infected plants , Ii , ( Eq . 6 ) , expressed as plants×m−2 , and of virus-free aphids , Xi , ( Eq . 7 ) and viruliferous aphids , Zi , ( Eq . 8 ) , expressed as aphids×plant−1 , i = ( 1 , 2 ) , and the model differs from the single host model ( Eq . 1–4 ) in allowing for aphids to visit Host 2 from Host 1 ( parameter Φ12 ) , from which acquisition of virus and transmission to Host 1 from Host 2 occurred ( parameter b21 ) . A similar set of equations describes the dynamics for Host 2 ( not shown ) ; the only differences would be the substitution subscript numbers 2 for 1 and vice versa . Definitions and values of model parameters are shown in Table 5 , in which Host 1 and Host 2 represent a short-lived and a long-lived genotype , respectively . Accordingly , model parameters were estimated from the experimental data obtained for Ler and Ll-0 , respectively , as in the previous sections . Because in Arabidopsis CMV infection is not lethal , virulence ( νi ) was estimated as the variation of the plant's lifespan by infection , DI , as compared with the lifespan of uninfected susceptible plants DS , following Day [51] . Mortality rates relate to lifespan by m = 1/DS , for non-infected plants , and ( m+νi ) = 1/DIi for infected plants . Data in Table 1 show that CMV infection does not affect the lifespan of Ler , while it delays the completion of Ll-0 life cycle , resulting in a negative virulence . Note that the lifespan of susceptible non-infected plants in Table 5 is shorter than the lifespan of mock-inoculated plants in Table 1 , as we considered that plants would not attract aphids until they had reached the four-leaf stage . Similarly , the mortality rate of aphids , α , was estimated as the inverse of their lifespan , which was of an average of 40 d in the conditions in which experiments were performed . Transmission probability by single aphids during each visit of an S plant from an I plant ( parameter b ) was according to the data in Table 2 . Following Madden et al . [50] we considered that the probability of acquisition of a non-persistently transmitted virus ( parameter a ) is the same as the probability of transmission , and that the aphid remains viruliferous for a maximum of 6 h ( parameter τ ) . The frequency of aphid visits to S plants from I plants was not estimated experimentally , and was given arbitrary values varying between 0 . 01 and 1 . These values may be realistic , because epidemiological studies of CMV in different regions of Spain for different years indicate transmission rates of 0 . 008–0 . 122 days−1 [52] . However , because mock inoculated and infected plants of genotypes Ler and Ll-0 showed different vector preference , the probability of aphid visits from infected Ll-0 to healthy Ler plants was considered to be 0 . 63 times the probability of visits from infected Ler to healthy Ler plants , infected Ll-0 to healthy Ll-0 or infected Ler to healthy Ll-0 plants , according to the data in Table 3 . For simulations , we considered a monomolecular growth of the population of susceptible plants , θ = rp ( K−T ) , where T = S+I , is the total plant population density , K is the maximum density of the population , and rp is its growth rate . To make the plant population constant during a growing season , we set rp = 1 to get a constant value of T = K . Simulations were performed for 5<K<500 , according to data on the variation of plant density in wild Arabidopsis populations in Central Spain over sites and years [29 , and unpublished results] . Similarly , we made the aphid population constant by considering its growth , ψ as monomolecular according to ψ = ra ( Q−P ) /K , where P = X+Z is the total aphid population , Q is its maximum density , and ra = 1 for a constant aphid population density per plant . Q was made to vary between 0 . 1 and 5 aphids per plant , which are realistic values of density of aphid populations in Central Spain [53] . For all simulations the initial condition was that 2% of the plants were CMV infected ( i . e . , values of I1 and I2 of 0 . 02Ki ) , according to the experimentally determined rates of vertical transmission of LS-CMV in Arabidopsis ( this work ) . All simulations were done in R .
For all model simulations the frequency of short-lived Host 1 and of the long-lived Host 2 in the plant population was made to vary between 0 and 100% . The threshold values of the density of the aphid population , Q , and of the probability of aphid visits to healthy plants from infected ones , Φ , necessary for the occurrence of a CMV epidemic , depended on the total plant population density ( K ) . For K = 5 epidemics occurred at Φ≥1 and Q≥3 , while for K = 50 and K = 100 epidemics occurred for Φ≥0 . 5 and Q≥1 , and for K = 250 or higher , epidemics occurred for Φ≥0 . 1 and Q≥3 , or for Φ≥0 . 5 and Q≥1 ( Fig . 1 and Table S1 ) . Hence , plant and aphid population densities are primary factors determining CMV infection . Note that although Φ was made to vary independently of Q , it is known that aphid mobility increases as aphid population density increases [50] , so that in a real situation Φ would not be independent of Q . The incidence of CMV ( i . e . , the fraction of infected plants , I ) at equilibrium was compared first for single-genotype populations of either the short-lived Host 1 or the long-lived Host 2 . It was found that incidence was higher for the long-lived host for a broad range of values of plant and aphid population density ( K , and Q values ) and of probability of aphid visits to non-infected plants from infected ones ( Φ values ) . Incidence was higher for the short-lived host only for plant and aphid densities , and probability of aphid visits to plants , that would result in low transmission rates , e . g . , for any Φ and Q value if K = 5 , and for low Φ and Q values for higher K , e . g . , Φ = 0 . 5 if K = 50 , Φ = 0 . 5 if Q = 1 when K = 100 , or Φ = 0 . 1 for Q = 5 or Q≥3 when K = 250 or K = 500 , respectively ( Fig . 1 and Table S1 ) . Hence , the highest competence and capacity to sustain aphid populations of the short-lived genotype is countered by the longer infectious period of the long-lived one when host and/or aphids are abundant and transmission is more effective . Then , CMV incidence at equilibrium was compared for populations that were built of mixtures of short- and long-lived hosts in different proportions . For heterogeneous host populations , model simulation analyses showed that for any values of plant and aphid population density , and of the probability of aphid visits to non-infected plants from infected ones , K , , Q , and Φ , the total incidence in the plant population , and the incidence for each plant genotype , was always higher than for single-host genotype populations ( Fig . 1 , and Table S1 ) . This result indicates that when the population is heterogeneous , there is a balance between the different epidemiological parameters , such as transmission rate , vector preference or infectious period , that result in higher incidence . However , for all explored conditions , the total incidence increased with increasing frequency of the short-lived host in the plant population , and incidence was always higher for the long-lived than for the short-lived host ( Fig . 1 and Table S1 ) . These two results indicate that the short-lived host was always a better source of infection for the long-lived one than vice versa . Thus , for all analyzed conditions , the short-lived host was a more efficient reservoir than the long-lived one . The difference between CMV incidence in the long-lived and short-lived hosts , however , depended on the density of the aphid population , Q , and their mobility , Φ , so that for any plant density value the difference increased with increasing Q if Φ≥1 , but had a minimum at Q = 3 for low Φ values ( e . g . , 0 . 5 ) . Thus , the efficiency as a reservoir of the short-lived host depended non-linearly on the rates of transmission ( Fig . 1 and Table S1 ) . The difference between CMV incidence in the long- and short-lived hosts also depended on the density and genetic composition of the host population: for K = 5 , the difference had relative maxima for different frequencies of the short-lived host according to the value of Q; for K = 50 , Φ = 0 . 5 and Q = 1 , the difference in CMV incidences in both hosts increased monotonically with the frequency of the short-lived host in the population . For any other values of K , Φ and Q , the difference between incidences in the long- and short-lived hosts decreased monotonically with the frequency of the short-lived host in the population ( see Fig . 2 for examples ) . Thus , the efficiency of the short-lived host as a reservoir depended in a complex non-linear way on the genetic composition of the host population , modulated by the density of the host and vector populations and by the rate of aphid mobility , hence modulated by the factors determining transmission efficiency . We considered next the values of CMV incidence in each host , and of their difference , after 150 iterations , which informs about the evolution of the epidemics before equilibrium . Also , this is a temporal frame more in line with the Arabidopsis life cycle of about 5 months . After 150 iterations the data ( Fig . 2 and Table S2 ) show a more dramatic variation of CMV incidence according to values of plant and aphid density and aphid mobility , and according to the frequency of the short-lived host in the population , than at equilibrium . Notably , the difference between incidences in the long- and short-lived hosts showed relative maxima for a wider range of values of plant and aphid density and aphid mobility . The difference could even show negative values , either with relative minima or monotonically increasing with increasing proportion of the short-lived host in the population , depending on the density of the host and vector populations . It is interesting to underscore that when plant population density was high ( high values of K ) , and both aphid population density and aphid mobility were low ( low values of Q and Φ ) , and the proportion of the short-lived host in the population was low , the incidence in the short-lived host was higher than in the long-lived one . This result indicates that under this set of conditions the epidemic progressed faster in the short-lived host at earlier times , and faster in the long-lived host at later times . Thus , this result underlies the role of the short-lived host as a reservoir .
Vectored viruses comprise a large fraction of emerging pathogens of humans , animals and plants , and important research efforts are devoted to understand the network of ecological and evolutionary factors that determine virus emergence [2]–[5] , [54] . Generally , emerging pathogens can be permanently maintained in reservoir hosts [2] , [3] , and identification of the determinants of host reservoir potential is central to understand emergence and infection dynamics . Within this context , the relationship between host lifespan and reservoir potential is an understudied field . It has been proposed that long-lived hosts will invest more in costly defenses against pathogens because they are more exposed to infection than short-lived hosts . Consequently , short-lived hosts will be more susceptible to infection , more competent as sources of infection and/or will sustain larger vector populations , hence being effective reservoirs for the infection of long-lived hosts [8] , [14] , [55] . Evidence for a link between host lifespan and reservoir potential derives mostly from the comparison of species within the host range of a pathogen [6] , [7] , [22] , [23] , [56]–[58] , and the underlying mechanisms have rarely been analyzed [e . g . 8 , 24 , 58] . To our knowledge , the relationship between host lifespan , defense and reservoir potential , has not been analyzed at the intra-species level , in spite of abundant evidence of genetic diversity resulting in polymorphisms for lifespan and defense within species . It has been shown that tolerance to CMV infection in A . thaliana correlates positively with lifespan , the longer-lived genotypes showing a higher tolerance to CMV infection [40] , [41] . Higher tolerance was , at least in part , due to the ability of long-lived genotypes to modify their developmental schedule upon infection , so that more resources were allocated to reproduction than to growth [40] , [41] . In this study we focus on two Arabidopsis genotypes that represent extremes of lifespan and tolerance , i . e . , Ler and Ll-0 [40] , [41] , as representatives of the short- and long-lived genotypes that co-exist in wild Arabidopsis populations in different regions of the Iberian Peninsula [29] , [32] , [35] . Ler and Ll-0 plants differed sharply in the temporal parameters of their life-cycle , in their morphology and in other relevant life-history traits ( Table 1 and [40] , [41] ) . Among these , it is noteworthy that seeds of the short-lived genotype , Ler , did not require a period of dormancy before germination , which will allow them to have more than one generation per growing season under field conditions [29] , [34] . This will not be the case for the long-lived Ll-0-like genotypes that may require a period of dormancy for germination and vernalization for flowering . CMV infection did not affect germination rate of any genotype tested , and CMV vertical transmission rate did not differ for the short- and long-lived genotypes . Thus , it is tempting to speculate that CMV will follow different strategies to increase its fitness in short- and long-lived genotypes . In the long-lived genotype , which has only one generation per year , infection increases the lifespan ( i . e . , the infectious period ) , which is not modified in the short-lived genotype , which may have more than one generation per year . Interestingly , in this system we cannot equate short and long lifespan with the physiological phenotypes of Quick- or Slow-Return , since rosette relative growth rate was higher in the long-lived Ll-0 genotype than in the short-lived Ler , but leaf mass per unit area did not differ between both genotypes . This unexpected result is at odds with evidence derived from the comparison of different plant species [8] , [24] , [59] , [60] , and suggests that the trade-off between lifespan and development and reproductive rates [61] may not apply at the level of within-species diversity , or at the temporal scale at which the lifespan of Ler and Ll-0 differ . The different performance of aphids in each genotype , regardless of similar LMA , could be explained by differences in nutrient composition in the phloem sap [62] , which would not translate into detectable differences in LMA . Using these two Arabidopsis genotypes we analyzed if lifespan , in addition to correlating positively with tolerance to CMV [41] , also correlated with susceptibility , competence and the ability to sustain vector populations; three parameters that determine reservoir potential [8] . Ler and Ll-0 plants were similarly competent sources of infection , as well as similarly susceptible to infection when transmission assays were performed between plants of the same genotype . These results agree with the non-significant difference in virus accumulation between Ler and Ll-0 plants , transmission rates of CMV in different hosts having been shown to increase with virus titer until saturation at high virus titers [63] , [64] . On the other hand , the susceptibility to infection of Ll-0 , but not of Ler , depended on the genotype of the plant from which the aphid vector acquired the virus . Hence , the susceptibility of the long-lived genotype will be , on the average , lower in a heterogeneous plant population than that of the short-lived genotype . The capacity to sustain aphid vector populations was higher in the short-lived Ler genotype than in the long-lived Ll-0 one , when both mock-inoculated and CMV-infected plants were considered . In addition , more aphids landed and settled , and more nymphs were generated , in CMV-infected , but not in mock-inoculated , Ler than in Ll-0 plants , indicating that the capacity to sustain vector populations was differentially modified for each genotype by virus infection , and was higher in the short-lived genotype . It is noteworthy that the results of our aphid host preference experiments agree with previous reports involving different host plant species infected with CMV , in which it was shown that after an initial attraction to CMV-infected plants , aphids migrated and settled into mock-inoculated ones at some time between 30 and 60 min after landing [65] , [66] . All these results taken together indicate that two of the three epidemiological parameters related to reservoir potential , susceptibility to infection and capacity to sustain vector populations , were higher in the short-lived than in the long-lived genotype , which would indicate a higher reservoir potential of the short-lived genotype . Thus , our results agree with predictions on the relationship between host lifespan and defense evolution [14] , [55] and with evidence derived from the comparison of different hosts of multi-host parasites infecting both plants and animals [6]–[8] , [56]–[68] . It is important to point-out that most previous evidence for a relationship between lifespan , defense and reservoir potential , was derived from the comparison of different species with large differences in lifespan , of the order of years , e . g . , rodent species with life expectancies of 2 vs . 4–8 years , or annual vs . perennial plant species [8] , [56] . Our results extend that evidence to a much narrower time-scale in lifespan differences , in the order of weeks , as expected for the variability in lifespan at the within species level for a short-lived plant . Although our experimental results derive from the study of only one short- and one long-lived genotype , they indicate that a link between lifespan and defense at the within-species level will result in polymorphisms for reservoir potential within a heterogeneous single-species host population . There is growing evidence for covariance between host traits determining reservoir potential and local extinction risk [55] . If longer lived , less susceptible and competent host genotypes are also at higher local extinction risk is a question to be explored , as it could determine both disease dynamics and the evolution of apparently unrelated traits in host populations . In terms of reservoir efficiency , the lower defenses , i . e . , higher susceptibility and capacity to sustain vector populations , of the short-lived Arabidopsis genotype could be balanced by the longer infectious period of the long-lived genotype , and the balance might be modulated by the demography and genetic composition of the host population . To explore how the efficiency as a reservoir is affected by host demography and by the genetic composition of the host population , we developed a simple epidemiological model which is essentially a simplification of that proposed by Madden et al . [50] with the addition of a virulence parameter to take into account the effect of infection on host plant mortality . The model was run for a set of realistic parameters derived from the above-discussed experiments with Ler and Ll-0 , and for a wide range of host plant and aphid population densities according to field observations [29] , [53] . The negative virulence in the long-lived Host 2 deserves some consideration . The negative virulence value result from the increase in lifespan of long-lived Arabidopsis genotypes upon infection , which associated with resource reallocation resulting in tolerance [41] , and is linked to a decrease in growth rate ( Table 1 ) . Under field conditions in which plants will compete , the slower-growing infected long-lived plants could be at a competitive disadvantage , hence suffering from a positive virulence as was shown in experiments in which plant density was manipulated [21] . Model simulation analyses were done considering different virulence values , the negative virulence value shown in Table 5 , virulence 0 , or a positive virulence of the same amount , with no substantial difference in the results . Thus , even in the case that negative virulence were an artifact of green-house experimentation , it will not affect the outcome of model simulation analyses . When the model was run for single-genotype populations of either short-lived or long-lived hosts , it was found that the short-lived host population sustained higher incidence of CMV only at low plant and/or aphid populations densities , i . e . , under conditions of low transmission rates . Thus , the higher reservoir potential of short-lived genotypes due to lower defense was partly countered by the longer infectious period of long-lived genotypes when transmission was highly efficient ( i . e . higher K , Q and Φ ) . In agreement with this result , we found that the equilibrium incidence of CMV was always higher for mixed-genotype host populations than for single-genotype populations of either short- or long-lived hosts , underscoring the role in determining the reservoir efficiency of both host defense and infectious period . The total CMV incidence in mixed-genotype populations increased with increasing frequency of the short-lived genotype , and incidence was always higher in the long-lived genotype subpopulation , indicating that for the realistic parameters and within the wide range of conditions considered , the short-lived genotype is always a better reservoir than the long-lived one . How good a reservoir was the short-lived genotype , however , depended in a complex way on the density of the host and aphid populations , as determinants of the probability of transmission . These non-linear effects were more pronounced when we analyzed the predictions of the models after 150 iterations , which may approximate the Arabidopsis growth period , rather than the equilibrium values . After 150 iterations , the difference between CMV incidence in long- and short-lived hosts showed maxima , or even became negative , for a given K value according to the Q and Φ values ( Fig . 2 and Table S2 ) . Data from 150 iterations also showed that the relative rate of epidemic growth in each host varied with time . The complex relationship found between efficiency as a reservoir of the short-lived genotype , the density of host and aphid populations , and the genetic composition of the host population , is in agreement both with model predictions on the variation of defense with host population traits [14] , [18] and with our experimental results with the same host-pathogen system [21] . In summary , in this work we show that the hypothesis stating that there is a correlation between host lifespan and investment in defenses against pathogens , developed and tested for the different hosts of multi-host pathogens , also holds for two genotypes of a single host species , which may differ in lifespan at much smaller temporal scales . Analyses of more short- and long-lived genotypes would be required for generalization , but our results indicate that the less efficient defenses of short-lived genotypes result in their higher reservoir potential . However , the reservoir potential of short-lived genotypes may be balanced in heterogeneous host populations by the longer infectious period of long-lived genotypes . Model simulations under realistic parameter ranges showed that this balance is modulated according to complex , non-linear relations , by the demography of both the host and vector populations , and by the genetic composition of the host population , an important conclusion that often is not considered in analyses of the evolutionary ecology of pathogen emergence . Thus , within-species genetic diversity for lifespan and defenses against pathogens will result in polymorphisms for pathogen reservoir potential , which will condition within-population infection dynamics . These results should be taken into account in the future in joint analyses of the population genetics of traits determining host defense and lifespan , to get a better understanding of the evolution of defense against pathogens in host populations . | Understanding pathogen emergence is a major goal of pathology , because of the high impact of emerging diseases . Pathogens emerge onto a new host from a reservoir , hence the relevance of identifying the determinants of host's reservoir potential . Host lifespan is considered as one such determinant: short-lived hosts will invest less in defenses , being more susceptible to infection , more competent as infection sources and/or will sustain larger vector populations , and thus , are effective reservoirs for long-lived host infection . Evidence for this hypothesis derives from analyses of different hosts of multihost pathogens , and here we examine whether it holds at the within-species level by comparing two genotypes of the plant Arabidopsis thaliana that differ in life-span and in tolerance to its natural pathogen Cucumber mosaic virus . Experiments showed that defenses to the virus and its aphid vector were less efficient in the short-lived genotype that , according to model simulations , was an effective reservoir under a large range of conditions . Reservoir potential , though , was modulated by the demography of host and vector and by the genetic composition of the host population . Thus , within-species genetic diversity for lifespan and pathogen defense will result in differences in reservoir potential , which will condition infection dynamics and host-pathogen co-evolution . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Experimental",
"Methods",
"Experimental",
"Results",
"Model",
"Methods",
"Model",
"Simulation",
"Results",
"Discussion"
] | [
"evolutionary",
"ecology",
"plant",
"science",
"plant",
"viral",
"pathogens",
"ecology",
"virology",
"plant",
"pathogens",
"plant",
"pathology",
"biology",
"and",
"life",
"sciences",
"microbiology",
"evolutionary",
"biology"
] | 2014 | The Relationship between Host Lifespan and Pathogen Reservoir Potential: An Analysis in the System Arabidopsis thaliana-Cucumber mosaic virus |
Hepatitis B virus ( HBV ) is one of the major etiological pathogens for liver cirrhosis and hepatocellular carcinoma . Chronic HBV infection is a key factor in these severe liver diseases . During infection , HBV forms a nuclear viral episome in the form of covalently closed circular DNA ( cccDNA ) . Current therapies are not able to efficiently eliminate cccDNA from infected hepatocytes . cccDNA is a master template for viral replication that is formed by the conversion of its precursor , relaxed circular DNA ( rcDNA ) . However , the host factors critical for cccDNA formation remain to be determined . Here , we assessed whether one potential host factor , flap structure-specific endonuclease 1 ( FEN1 ) , is involved in cleavage of the flap-like structure in rcDNA . In a cell culture HBV model ( Hep38 . 7-Tet ) , expression and activity of FEN1 were reduced by siRNA , shRNA , CRISPR/Cas9-mediated genome editing , and a FEN1 inhibitor . These reductions in FEN1 expression and activity did not affect nucleocapsid DNA ( NC-DNA ) production , but did reduce cccDNA levels in Hep38 . 7-Tet cells . Exogenous overexpression of wild-type FEN1 rescued the reduced cccDNA production in FEN1-depleted Hep38 . 7-Tet cells . Anti-FEN1 immunoprecipitation revealed the binding of FEN1 to HBV DNA . An in vitro FEN activity assay demonstrated cleavage of 5′-flap from a synthesized HBV DNA substrate . Furthermore , cccDNA was generated in vitro when purified rcDNA was incubated with recombinant FEN1 , DNA polymerase , and DNA ligase . Importantly , FEN1 was required for the in vitro cccDNA formation assay . These results demonstrate that FEN1 is involved in HBV cccDNA formation in cell culture system , and that FEN1 , DNA polymerase , and ligase activities are sufficient to convert rcDNA into cccDNA in vitro .
Hepatitis B virus ( HBV ) is a major pathogenic cause of human cirrhosis and hepatocellular carcinoma [1] . Infectious HBV particles contain relaxed circular DNA ( rcDNA ) encapsidated by core proteins [2] . After entering the host hepatocyte , rcDNA is converted into covalently closed circular DNA ( cccDNA ) , which is stably maintained as an episome in the nucleus . cccDNA serves as the template for all HBV transcripts , including pregenomic RNA ( pgRNA ) , a viral replicative intermediate [2–4] . pgRNA , viral reverse transcriptase P protein , and core proteins assemble into a nucleocapsid , where pgRNA undergoes reverse transcription by the P protein to produce rcDNA . The mature nucleocapsid is further assembled with surface proteins to allow secretion as an infectious virion . Alternatively , the rcDNA containing nucleocapsid is recycled back to the nucleus to maintain the pool of cccDNA [5] . Reverse-transcriptase inhibitors are the major medical intervention for controlling HBV infection . These inhibitors can effectively shut down viral replication , but are unable to eliminate cccDNA from infected hepatocytes; this inability often leads to viral rebound upon therapy withdrawal [2 , 3 , 6] . New therapeutic approaches are needed to target the mechanisms of cccDNA maintenance and generation . However , a lack of comprehensive knowledge on the molecular mechanisms of cccDNA formation and maintenance has hampered the effective development of such approaches . The cccDNA precursor rcDNA has unique structural features that are absent from cccDNA . These include a P protein-linked sequence approximately 10 nucleotides in length , known as terminal redundancy ( r ) , which is located at the 5′ end of the minus-strand DNA , and a small RNA oligomer attached at the 5′ end of the plus strand [2 , 6] . The first step in cccDNA conversion from rcDNA is removal of the P protein and RNA oligomer linkage from the 5′ ends . Resulting protein-free rcDNA or deproteinized rcDNA is proposed to be a direct precursor to cccDNA [7 , 8] . In addition to removing the r sequence and RNA oligomer from rcDNA , filling-in the single-stranded region and ligation of nicks in both DNA strands are required for cccDNA formation . Flap endonuclease 1 ( FEN1 ) is a flap structure-specific endonuclease . FEN1 plays a role in removing 5′-flap structures formed during Okazaki fragment maturation and long-patch base excision repair ( LP-BER ) [9 , 10] . Because the r sequence and RNA oligomer at the 5′ end of rcDNA may form a 5′-flap structure , we examined the possible involvement of FEN1 in the removal of 5′-flap structures from rcDNA and its subsequent conversion to cccDNA .
To determine whether FEN1 protein removes the r sequence from rcDNA , we designed a synthetic DNA substrate that mimics the 5′-flap structure of the r sequence ( S1A Fig ) by modifying an established FEN assay [11] . Human wild-type ( wt ) and catalytic mutant FEN1 proteins were prepared by immunoprecipitation ( S1B Fig ) and used for the HBV-FEN assay . Cleavage of the r sequence was determined by fluorescence intensity ( S1C Fig ) and polyacrylamide gel electrophoresis ( PAGE ) ( S1D Fig ) . Incubation with immunoprecipitated FEN1 protein caused an increase in cleavage of the synthetic r sequence over time when compared with that of the mock-precipitated protein ( S1C Fig ) . Conversely , two catalytic mutant FEN1 proteins [12] lost their cleavage activity ( S1D Fig ) . Previous studies demonstrated the inhibition of flap endonuclease activity by the FEN1 inhibitor 3-hydroxy-5-methyl-1-phenylthieno[2 , 3-d]pyrimidine-2 , 4 ( 1H , 3H ) -dione ( PTPD ) [11 , 13] . In the current study , we examined whether PTPD could inhibit FEN activity of the immunoprecipitated FEN1 protein by using the HBV FEN assay . PTPD addition strongly inhibited FEN1 cleavage activity ( S1E Fig ) . We used this inhibitor to explore the possible involvement of FEN1 in cccDNA formation in a cell culture system . Hep38 . 7-Tet cells replicate HBV and accumulate cccDNA after removing tetracycline from the culture medium [14 , 15] . Using Hep38 . 7-Tet cells , the production of viral HBV intermediates was determined ( Fig 1A–1E ) . Hirt extraction of cccDNA was followed by T5 exonuclease treatment to digest non-cccDNA molecules . T5 exonuclease removes nucleotides from 5′ termini , at gaps and nicks of linear or circular double-stranded DNA . The levels of cccDNA were determined by cccDNA-selective qPCR , which targets the gap region in rcDNA [16 , 17] ( Fig 1C ) . To demonstrate selective detection of cccDNA by the cccDNA-selective qPCR , a control experiment was performed . The Hirt extracted DNA and secreted HBV DNA were prepared from Hep38 . 7-Tet cells . The same copy number of the Hirt extracted DNA and secreted HBV DNA was applied to the cccDNA-selective qPCR . Our cccDNA-selective qPCR quantitatively detected cccDNA only from the Hirt-extracted HBV DNA , but not from secreted HBV DNA ( S2 Fig ) . To characterize the effect of PTPD treatment in Hep38 . 7-Tet cells , the effect of a reverse-transcriptase inhibitor , 3TC , was compared with that of PTPD . 3TC suppressed secreted and cytoplasmic nucleocapsid-associated DNA ( cytoplasmic NC-DNA ) and cccDNA levels ( Fig 1A–1C ) . These results were expected , as HBV NC-DNA and cccDNA generation were completely dependent on reverse transcription in Hep38 . 7-Tet cells ( S3 Fig ) , and 3TC was simultaneously added when Tet-CMV promoter was activated by removal of tetracycline from culture medium . Pre-C mRNA is transcribed from cccDNA , but not from the HBV transgene chromosomally integrated in cellular genome in the Hep38 . 7-Tet cells [18] . Consistent with the decreasing cccDNA levels in 3TC-treated Hep38 . 7-Tet cells , 3TC also reduced pre-C mRNA levels ( Fig 1C and 1E ) . Conversely , PTPD significantly decreased both cccDNA and pre-C mRNA levels ( Fig 1C and 1E ) . Importantly , PTPD did not affect the levels of secreted and cytoplasmic NC-DNAs ( Fig 1A and 1B ) , confirming that transcription of pgRNA from the chromosomal copy was not affected by PTPD . Treating Hep38 . 7-Tet cells with PTPD ( 5 μM ) for 5 days did not affect cellular proliferation ( S4 Fig ) . These results suggest that PTPD blocked a step of cccDNA formation but did not inhibit reverse-transcription of pgRNA and cellular proliferation , as well as transcription from the HBV transgene . The FEN1 inhibitor experiments suggested that FEN1 is involved in cccDNA formation , but not rcDNA formation . However , the observed reduction of cccDNA level by PTPD was moderate , compared to that of 3TC ( Fig 1C ) . To confirm this result by different approaches , we performed small interfering RNA ( siRNA ) -based knockdown and genome editing . Two siRNAs designed against FEN1 mRNA and control siRNA were transfected into Hep38 . 7-Tet cells . Transfection of FEN1 siRNA reduced FEN1 mRNA and protein expression by more than half of the control siRNA levels ( Fig 2A ) . Consistent with the result in Fig 1 , knockdown of FEN1 expression reduced , albeit moderately , the cccDNA levels without affecting the cytoplasmic NC-DNA level generated from the HBV transgene ( Fig 2B and 2C ) . To further confirm the requirement for FEN1 in cccDNA formation , CRISPR/Cas9-mediated genome editing was applied to Hep38 . 7-Tet cells . We obtained two independent lines of FEN1+/− Hep38 . 7-Tet cells; each had one base ( T ) insertion in exon 2 of the FEN1 gene . The one-base insertion caused a frame shift and premature stop codon at amino acid position 102 , immediately after insertion ( S5 Fig ) . RT-qPCR and Western blot analyses demonstrated reduced FEN1 expression up to approximately half of the parental Hep38 . 7-Tet cells ( Fig 2E ) . Consistent with the results obtained from the knockdown experiments , FEN1+/− Hep38 . 7-Tet cells produced cccDNA at approximately half the level of parental Hep38 . 7-Tet cells ( Fig 2F ) . Southern blotting also showed moderately reduced cccDNA levels ( Fig 2G , right ) , while intact cytoplasmic rcDNA production was observed in FEN1+/− Hep38 . 7-Tet cells ( Fig 2G , left ) . Taken together , the knockdown and genome editing results clearly demonstrated that reduced FEN1 expression decreased cccDNA levels without reducing cytoplasmic rcDNA levels in Hep38 . 7-Tet cells . Recent studies documented successful HBV infection in NTCP-expressing HepG2 cells [19 , 20] . Thus , we examined the involvement of FEN1 in cccDNA formation using NTCP-expressing HepG2 ( HepG2-hNTCP-C4 ) cells [19] . HepG2-hNTCP-C4 cells were pretreated with PTPD for 1 day , and subsequently infected with HBV . HBV-infected HepG2-hNTCP-C4 cells were cultivated for 3 days in the presence of PTPD , and cccDNA levels were determined by Southern blotting . As indicated in Fig 3A , the cccDNA level was mildly reduced by PTPD treatment ( 51 . 3% of control cccDNA level ) . Exposure of PTPD for 7 days in this cell line did not affect cellular proliferation ( S6 Fig ) . To further confirm the results of infected HepG2-hNTCP-C4 , we used PXB primary human hepatocytes derived from liver-humanized mice [21] . As shown in Fig 3B , PTPD treatment both inhibited secretion of HBV DNA and reduced HBV RNA levels in HBV-infected PXB cells . On the other hand , 3TC suppressed HBV DNA secretion but did not reduce HBV RNA levels . Importantly , in the infection model , rcDNA in the inoculum is first converted into cccDNA in the nucleus , and the newly produced cccDNA is transcribed into pgRNA , resulting in encapsidation and reverse-transcription in the nucleocapsid , yielding the mature virion ( S3 Fig ) . Therefore , it is reasonable that PTPD treatment reduced both cccDNA formation and HBV DNA secretion in HBV-infected cells ( Fig 3A and 3B ) . These results indicated that the FEN1 inhibitor blocks cccDNA formation following viral replication in infected NTCP-HepG2 cells and human primary hepatocytes . It was previously reported that a point mutation ( D181A ) of FEN1 results in a loss of nuclease activity , while deletion of 20 amino acids from the C-terminus ( ΔC ) of FEN1 results in a loss of binding to the telomere maintenance protein , WRN , and truncation of nuclear localization signal ( Fig 4A ) [22–24] . We first tested the FEN activity of these mutants with the HBV-FEN assay used in Fig 1 . As demonstrated in Fig 4B , FEN1 wt protein cleaved the r sequence , while FEN1 ΔC showed lower cleavage activity and D181A could not cleave r . Importantly , all protein levels were comparable ( S7 Fig ) . To assess the requirement of nuclease activity and the C-terminus of FEN1 for cccDNA formation , we knocked-down endogenous FEN1 expression and simultaneously overexpressed either FLAG-tagged-wt , D181A , or ΔC FEN1 protein using the pResQ vector [25] . The pResQ lentiviral expression vector simultaneously expresses both short hairpin RNA ( shRNA ) targeting the 3′-untranslated region ( UTR ) of FEN1 mRNA ( shFEN1 ) and exogenous wt or mutant FEN1 protein ( S8A Fig ) . As shown in S8B Fig , endogenous FEN1 expression in shFEN1-transduced cells ( mock , wt , D181A , ΔC ) was significantly lower than in cells transduced with control shRNA ( shCtrl ) . Furthermore , overall FEN1 expression levels in shFEN1-expressing wt , D181A , and ΔC transfectants were substantially higher than in shCtrl- and shFEN1-expressing mock transfectants , due to the exogenous expression of FEN1 protein . Western blotting confirmed the reduction of endogenous FEN1 and the expression of exogenous FEN1 protein , although endogenous FEN1 protein is visible in knockdown cells ( S8C Fig ) . Cytoplasmic NC-DNA and cccDNA levels were also determined in these transfectants . All cells produced cytoplasmic NC-DNA at similar levels , while shFEN1-mock transfectants exhibited lower levels of cccDNA that were restored by wt FEN1 expression ( Fig 4C ) . In addition , D181A and ΔC mutant transfectants tended to exhibit lower cccDNA levels than that of wt , although this tendency was not statistically significant ( Fig 4C ) . Although these experiments do not conclusively demonstrate cccDNA formation roles for the FEN1 protein catalytic site and C terminus , they do show that FEN1 expression is required for cccDNA formation . Since the reduction of the cccDNA level in FEN1+/− Hep38 . 7-Tet cells was moderate ( Fig 2G ) , we further examined the additive effect of genome editing and shRNA knockdown on cccDNA formation . pResQ shFEN1-mock lentiviral vector was transduced into two independent clones ( #1 and #2 ) of FEN1+/− Hep38 . 7-Tet cells . Endogenous FEN1 protein was effectively reduced in these FEN1+/− shFEN1 cells ( S8D Fig ) . Southern blotting analysis ( Fig 4D ) showed that FEN1+/− shFEN1 cells produced cytoplasmic NC-DNA as effectively as shRNA control cells ( shCtrl ) . Meanwhile , the cccDNA level was clearly reduced in FEN1+/− shFEN1 cells compared with shRNA control cells ( 39 . 7 and 25 . 9% of the control cccDNA level , respectively ) . The additive effect of cccDNA reduction with the combination of FEN1+/−and FEN1 shRNA clearly indicates the requirement of FEN1 for cccDNA formation . It has been reported that rcDNA in Hirt extraction was reduced upon inhibition of cccDNA formation by knock-out of DNA ligases LIG1 and LIG3 [26] . It was proposed that nicked cccDNA behaves similar to rcDNA during electrophoresis , and a concurrent decrease of rcDNA may be due to a decrease in cccDNA formation . Reduction of rcDNA in Hirt DNA is also observed in our study ( Figs 2G and 4D ) . Subcellular localization of FEN1 protein was examined in HBV-replicating Hep38 . 7-Tet cells . As expected [25] , wt FEN1 protein localized to the nucleus , which was disrupted by the ΔC mutation ( Fig 5A ) . We next utilized immunoprecipitation in order to determine whether FEN1 can associate with HBV DNA . c-Myc-tagged-wt or ΔC FEN1-expressing Hep38 . 7-Tet cells ( Fig 5B ) were treated with formaldehyde to cross-link protein and DNA , and FEN1 proteins were immunoprecipitated using a c-Myc antibody . The cross linkage and fragmentation of DNA that were necessary for this approach can make it difficult to judge which viral DNA forms are precipitated during FEN1 immunoprecipitation . However , the ability of FEN1 to associate with any of the viral DNAs can be estimated by comparison of the immunoprecipitated HBV DNA levels in the FEN1 with the control IgG conditions . As predicted , a significantly higher level of HBV DNA was detected in the FEN1 wt precipitate compared with that in the control as well as with the ΔC mutant precipitation ( Fig 5C ) . Importantly , the ΔC mutant , missing its nuclear targeting ability , exhibited decreased HBV DNA binding , relative to the wild type FEN1 protein . This finding suggests that wild type FEN1 localizes to the nucleus and associates with nuclear HBV DNA , such as nuclear rcDNA , either directly or indirectly . Because FEN1 protein can remove the HBV r sequence in vitro ( Fig 1A ) and cellular experiments suggest a role of FEN1 in cccDNA formation ( Figs 1–4 ) , we assessed whether FEN1 can participate in any process of conversion of rcDNA to cccDNA in vitro . First , the FEN activity of recombinant FEN1 protein was reconfirmed by the HBV-FEN assay . Consistent with the results in Fig 1 , recombinant FEN1 protein cleaved the r sequence from the synthetic HBV substrate in a dose-dependent manner ( S9 Fig ) . Next , we determined whether recombinant enzymes , including FEN1 , could convert the purified rcDNA into cccDNA . The purified NC-DNA from Hep38 . 7-Tet cells was incubated with recombinant FEN1 , DNA polymerase , and DNA ligase , and cccDNA formation was determined by cccDNA selective-qPCR , rolling circle amplification ( RCA ) , and Southern blot ( Fig 6A–6D ) . RCA is able to speficically amplify closed circular DNA . The combination of FEN1 , DNA polymerase , and DNA ligase led to the significant production of cccDNA ( Fig 6B–6D ) . Meanwhile , incubation with two enzymes ( DNA polymerase and DNA ligase ) did not support efficient cccDNA formation ( Fig 6B–6D ) . DNA sequencing of the closed circular DNA produced by three enzymes confirmed that the rcDNA gap region was precisely filled and did not have any mutations ( S10 Fig ) . Furthermore , replication competency of in vitro-generated cccDNA was confirmed by transfecting the self-circularized RCA product into HepG2 cells ( Fig 6E and 6F ) . These results indicate that the circular DNA generated by incubating with FEN1 , DNA polymerase , and DNA ligase is functional HBV cccDNA .
Host DNA repair factors are expected to be involved in cccDNA conversion because the virus genome does not encode the responsible DNA modifiers [2 , 3 , 6 , 27] . The TDP2 enzyme has been proposed to remove P protein [28] , although another study reported that TDP2 is not required for cccDNA formation in vivo [29] . We previously showed that the host DNA repair enzyme UNG removes uracil residues from deaminated duck HBV ( DHBV ) cccDNA ( or its precursor ) , thus changing its mutation frequency [30] . FEN1 plays a role in various DNA metabolic pathways , including Okazaki fragment maturation and LP-BER . During lagging strand DNA synthesis , Polδ/Polε use the strand exchange activity to produce the 5′-flap structure , and then FEN1 cleaves the 5′-flap and generates a ligatable end to facilitate lagging strand synthesis . In LP-BER , FEN1 removes the 5′-flap structure containing a damaged sugar and generates a ligatable 5′ end to facilitate its repair process [9 , 10] . However , it remains unknown whether other back-up systems can substitute for absence of FEN1 activity . During cccDNA formation , rcDNA-specific structures have to be removed . We assume that some of these rcDNA-specific structures form the 5′-flap structure . FEN1 is a good candidate to remove them . To test this , we utilized FEN1 loss-of-functional approaches in HBV-replicating cells , including a FEN1 inhibitor ( Figs 1 and 3 ) , siRNA and shRNA knockdown ( Figs 2 and 4 ) , and CRISPR/Cas9-mediated genome editing ( Fig 2 ) . The four different approaches of FEN1 loss of function showed the same trend , that is , a moderate reduction in cccDNA levels . We also utilized a combined approach of the genome editing and shRNA knockdown ( Fig 4D ) . This method resulted in FEN1+/−-FEN1 shRNA Hep38 . 7-Tet cells with a clearly reduced cccDNA level . This reduction seemed to be specific to cccDNA because NC-DNA production was not reduced in Hep38 . 7-Tet cells , which could produce NC-DNA from genome integrated HBV transgene . We interpreted this reduced level of cccDNA as a specific phenotype of FEN1 loss-of-function , rather than off-target effects from each approach . Infection experiments showed that inhibition of FEN1 activity reduced HBV DNA secretion in NTCP-expressing cells and primary human hepatocytes ( Fig 3 ) . FEN1 protein could cleave the r sequence in vitro ( Figs 4 and S1 ) and convert purified rcDNA into cccDNA along with DNA polymerase and ligase in vitro ( Fig 6 ) . Altogether , these results demonstrate , for the first time , that the host DNA repair factor FEN1 is involved in HBV cccDNA formation , at least in the experimental models used in this study . The cccDNA was not eliminated completely , even when the combination approach for FEN1 loss-of-function was employed , suggesting the possibility of other redundant enzymes . FEN1 is a member of the RAD2/XPG structure-specific 5′-nuclease family [31 , 32] . Among them , exonuclease 1 ( Exo1 ) is another candidate to remove the HBV r sequence from rcDNA , because it has both 5′ to 3′ exonuclease activity and endonuclease activity of the 5′-flap structure . Moreover , yeast genetic studies suggested that Exo1 and FEN1 activities may have a redundant role [33 , 34] . The human genome encodes another flap-structure specific endonuclease designated DNA2 . DNA2 plays a role to resolve a flap structure during Okazaki fragment maturation in yeast [35] . Further studies are needed to determine the host players other than FEN1 that remove the flap structure from rcDNA . Inhibition of FEN1 activity did not lead to an obvious reduction in proliferation , at least in the experimental conditions used in this study ( S4 and S6 Figs ) . Consistent with our observation , it was reported that the PTPD inhibitor showed little or no effect on cell growth of the T24 bladder cell line , but increased sensitivity to a DNA damage agent , i . e . methyl methanesulfonate [13] . Moreover , FEN1 mutations that abrogate nuclease activity have been detected in lung cancers and corresponding knock-in mice are viable with autoimmune , chronic inflammatory , and cancer phenotypes [36] . Meanwhile another knock-in mice of FEN1 mutant ( F343A and F344A ) that lose ability to bind PCNA but retain nuclease activity , die at birth [37] . On the other hand , FEN1−/−mice , which have a complete knock-out of FEN1 , have a lethality phenotype as early as embryonic day 3 . 5 [38] . Recent biochemical and genetic approaches identified more than 30 FEN1 associating proteins [32] including proteins involving in DNA replication , such as PCNA , apoptosis , telomere stability , post-transcriptional modification , and DNA repair . It is likely that complete loss of FEN1 protein in mammal manifests as a disturbance of cellular survival because both nuclease-dependent and -independent functions of FEN1 are disrupted . Meanwhile , inhibition of nuclease activity of FEN1 may not result in immediate disturbance of cellular proliferation . Consistent with this idea , the FEN1−/− cell line was not established in this study , even by CRISPR/Cas9-mediated genome editing . The HBV-FEN assay revealed that FEN1 could remove the r sequence from a synthetic HBV DNA flap substrate . Moreover , the combination of FEN1 , DNA polymerase , and DNA ligase was sufficient to convert cccDNA from purified rcDNA in vitro . It has been reported that Polδ and Lig I cooperate with FEN1 in Okazaki fragment maturation , and Polβ and Lig III cooperate with FEN1 in LP-BER [10 , 32] . However , the specific polymerase and ligase involved in HBV cccDNA formation remain unknown . Interestingly , the T5 exonuclease-resistant cccDNA ( Fig 6C , top ) migrated at approximately 3 . 4 kbp which is much higher position than that of cccDNA formed in infected hepatocytes , suggesting that its topology was different from cellular cccDNA . It is also possible for other additional factors such as topoisomerase and gyrase to be involved in cccDNA formation in vivo . Thus , further studies need to determine other host factors responsible for cccDNA formation . In summary , we demonstrate that reduced FEN1 expression and activity decreases cccDNA levels , and that FEN1 protein can bind and cleave the 5′-flap structure of HBV rcDNA in vitro to facilitate cccDNA conversion . The data implicate FEN1 as a critical enzyme involved in HBV cccDNA formation .
Hep38 . 7-Tet cells derived from the HepAD38 cell line ( obtained from Dr . Christoph Seeger at Fox Chase Cancer Center , Philadelphia ) [15] , and HepG2-hNTCP-C4 cells derived from HepG2 cells ( obtained from the JCRB Cell Bank ) [19] were cultured as described previously . Hep38 . 7-Tet cells were cultured with 0 . 3 μg/ml tetracycline to terminate HBV transcription . HBV production was induced in the cells by incubation in a tetracycline-free medium . 293FT cells ( purchased from Invitrogen ) were cultured as described previously [39] . PXB primary human hepatocytes were derived from liver-humanized mice [21] . The culture medium was purchased from PhoenixBio . For FEN1 inhibition experiments , PTPD ( 3-hydroxy-5-methyl-1-phenylthieno[2 , 3-d]pyrimidine-2 , 4 ( 1H , 3H ) -dione; Glixx Laboratories ) [13] was added to the culture medium . The HBV-FEN assay was performed as described previously [11] with minor modifications . Wild-type ( wt ) and mutant human FEN1 proteins were produced by transfecting FEN1 expression vectors [12] ( S1 Table ) into 293FT cells and enriched by immunoprecipitation with anti-c-Myc agarose affinity gel ( A7470; Sigma-Aldrich ) , as described below . The DNA substrate was prepared by annealing of “flap , ” “quencher , ” and “template” oligonucleotides containing the HBV sequences listed in S2 Table ( also see S1A Fig ) . Since 5-carboxytetramethylrhodamine ( TAMRA ) is attached to the 5′ end of the r sequence corresponding to the flap oligonucleotide , cleavage of the flap oligonucleotide by FEN activity can be measured as increasing fluorescence . DNA substrates were incubated with either FEN1 immunoprecipitants at room temperature or recombinant protein Thermostable FEN1 ( Thermococcus species 9°N origin; New England Biolabs ) at 65°C . Kinetic fluorescence data were collected on PowerScan ( DS Pharma Biomedical ) . Cleavage of the labeled , “flap” oligonucleotide was confirmed with 6 M urea/20% polyacrylamide gel electrophoresis ( S1D Fig ) . Purification of HBV DNA ( supernatant , cytoplasmic NC-DNA , and cccDNA ) and total RNA were described previously [30] with minor modification . HBV DNA in culture supernatant was extracted using a NucleoSpin kit ( Takara ) according to the manufacturer’s protocol . The purified HBV DNA from this fraction is designated as secreted HBV in this study . Viral DNAs from enveloped virions and naked capsids are included in this fraction . For cytoplasmic NC-DNA , the cells were lysed with buffer [10 mM Tris-HCl ( pH 8 . 0 ) , 1 mM EDTA , 1% NP-40 , 8% sucrose , proteinase inhibitor cocktail ( Roche ) ] . After centrifugation , supernatants were collected and further treated with DNase I and RNase A . NCs were then digested with proteinase K and sodium dodecyl sulfate ( SDS ) . The cccDNA purification was performed using a modified Hirt extraction procedure [30] . The Hirt-extracted DNA was purified and treated with T5 exonuclease ( New England Biolabs ) to digest linear and open circular DNA according to the manufacturer’s instructions . Total RNA was treated with amplification grade DNase I ( Thermo Fisher Scientific ) and reverse transcribed using an oligo ( dT ) primer and the SuperScript III kit ( Thermo Fisher Scientific ) . qPCR analysis of resulting cDNA was performed using SYBR green ROX ( Toyobo ) with MX3000 ( Stratagene ) as described previously [40] . For cccDNA quantification , TaqMan probe and cccDNA-selective primers spanning the gap region of rcDNA were used [16 , 17] . Validation of selective amplification for cccDNA but not rcDNA , is shown in S2 Fig . Primers and probe sequences are listed in S2 Table . RCA was performed as described previously [30 , 41] using purified DNA from the Hirt extraction . In brief , the DNA ( T5 exonuclease treated ) was mixed with 8 HBV-specific primers ( S2 Table ) , denatured at 95°C for 3 min; cooled sequentially at 50°C for 15 s , 37°C for 15 s , and room temperature , and reacted with the phi29 DNA polymerase ( New England Biolabs ) at 37°C for 16 h . RCA products were digested with EcoRI , which cuts HBV cccDNA once . The digested products were analyzed by gel electrophoresis and ethidium bromide staining . Immunoprecipitation and Western blotting were performed as described previously [30 , 40] . To analyze FEN1-HBV DNA binding , cells were fixed with 1% formaldehyde for 10 min at room temperature , quenched with 125 mM glycine , resuspended in lysis buffer ( 50 mM Tris-HCl pH 8 . 0 , 5 mM EDTA , 150 mM NaCl , 1% Nonidet P-40 , 0 . 1% sodium dodecyl sulfate [SDS] , protease inhibitor ) , sonicated in a Bioruptor sonication device ( Diagenode ) for 10 min using pulses of 30 s , and immunoprecipitated with anti-c-Myc antibody ( 9E10 , sc-40; Santa Cruz Biotechnology ) and protein G Sepharose ( GE Healthcare ) overnight at 4°C . Following proteinase K/SDS digestion , DNA was extracted with phenol/chloroform and precipitated with ethanol . Target DNA fragments were analyzed by qPCR as described above . The antibodies used for Western blotting were: mouse anti-FEN1 ( 4E7 , GTX70185 , GeneTex ) , rabbit anti-glyceraldehyde-3-phosphate dehydrogenase ( GAPDH ) ( G9545; Sigma-Aldrich ) , mouse anti-FLAG ( F3165 , Sigma-Aldrich ) , mouse anti-c-Myc ( 9E10 , sc-40; Santa Cruz Biotechnology ) , anti-rabbit Igs-horseradish peroxidase ( HRP ) ( ALI3404; Biosource ) , and anti-rabbit/anti-mouse IgG-HRP TrueBlot ( 18–8816 and 18–8877; eBioscience ) . Southern blotting was performed as described previously [30] . HBV DNAs were detected using a probe spanning the entire viral genome . Rhamda DNA probe was also simulateiously added to hybridzation buffer to visualize DNA size marker . Probe labeling and signal development was performed using the AlkPhos direct labeling system ( Amersham ) , and the signals were detected using the LAS1000 imager system ( Fuji Film ) . The Hirt-extracted DNAs were heat denatured at 95°C for 10 min and then subjected to EcoRI digestion to linearize DNAs . For the in vitro cccDNA formation ( Fig 6C ) , HBV DNAs were treated with T5 exonuclease ( New England Biolabs ) to eliminate any DNAs , except double-stranded closed circular DNA . After phenol-chloroform extraction , DNAs were digested with EcoRI , and then agarose gel electrophoresis was performed . Cell viability was evaluated using the Premix WST-1 Cell Proliferation Assay System ( Takara ) according to the manufacturer's instructions . The cell lines used for the WST-1 assay were sensitive to puromycin; hence , puromycin was used as a control . Two FEN1-specific siRNAs and control siRNA were purchased from Santa Cruz Biotechnology ( sc-37795 , sc-37007 ) and Sigma-Aldrich ( SASI_Hs02_00336939 ) . Lipofectamine 3000 ( Thermo Fisher Scientific ) was used to perform transfections with these siRNAs according to the manufacturer’s instructions . Cells and viruses were analyzed 4 days after transfection . Human FEN1-targeting oligonucleotides ( target sequence with the protospacer adjacent motif is in exon 2: 5′-AGCTGGCCAAACGCAGTGAGCGG-3′ ) were designed and cloned into the BbsI site of the pX330-U6-Chimeric_BB-CBh-hSpCas9 vector ( a gift from Feng Zhang , Addgene plasmid # 42230 ) [42] . The resulting pX330-FEN1 vector was co-transfected into Hep38 . 7-Tet cells with pIRES-GFP-bsd , a blasticidin-resistant gene expression vector ( S1 Table ) . Transfected clones were then selected using limiting dilution in the presence of blasticidin , and genome editing was confirmed by direct sequencing of the targeted region ( oligonucleotides are listed in S2 Table ) . Lentivirus-mediated gene transduction was performed as described previously [40] , using pResQ shFEN3 3XF-FEN1 wt , pResQ shFEN3 3XF-FEN1 D181A , and pResQ shFEN3 3XF-FEN1 ΔC ( gifts from Sheila Stewart , Addgene plasmid # 17752 , 17753 , and 17754 , respectively ) [25] . Construction of pResQ vectors is described in S7A Fig and S1 Table . HBV infection was performed as described previously [19 , 21] . Briefly , HBV ( genotype D ) was prepared from the culture supernatant of Hep38 . 7-Tet cells and concentrated with PEG8000 precipitation . The amount of HBV DNA was quantified by qPCR as described above . HepG2-hNTCP-C4 cells and PXB cells were seeded in collagen-coated plates , and the medium was replaced with fresh medium containing 4% PEG8000 and the prepared HBV ( 15 , 000 GE/cell for HepG2-hNTCP-C4 infection , 100 GE/cell for PXB infection ) . Twenty-four hours post-infection , the infected cells were washed three times with phosphate buffered saline and switched to fresh medium with PTPD or lamivudine ( 3TC ) . The cells and culture supernatants were collected at the indicated days post infection ( d . p . i . ) . Purified NC-DNA from Hep38 . 7-Tet cells was used as substrate DNA . NC-DNA ( 108 copies ) was incubated with 32 units ( U ) of Thermostable FEN1 in ThermoPol Buffer ( New England Biolabs ) at 65°C for 10 min , followed by incubation with 8 U of Bst DNA polymerase , 40 U of Taq DNA ligase , 100 μM dNTPs , and NAD+ ( all from New England Biolabs ) . After further incubation at 37°C for 20 min , DNA was purified by phenol/chloroform extraction and ethanol precipitation , and subjected to cccDNA-selective qPCR or RCA , as described above . The sequence corresponding to gap region of rcDNA was confirmed by direct sequencing ( oligonucleotide is listed in S2 Table ) . To verify the replication competence of resulting products , EcoRI-digested RCA products ( 3 . 2 kb ) were extracted from the gel; 50 ng of these fragments were subjected to self-circularization by T4 DNA ligase ( Takara ) . For the negative control , HBV plasmid ( pPB [30] ) was amplified with RCA and then digested with PstI . Because the 5 . 4-kb PstI fragment has a partial HBV sequence , it was used as a replication-defective control . Self-circularized DNAs were transfected into HepG2 cells . Three days after transfection , HBV DNAs were analyzed by qPCR . Statistical analyses were performed using GraphPad Prism ( GraphPad Software ) . Significance between two groups was determined using a Student’s t-test , while significance between three or more groups was determined using a one-way ANOVA with Dunnett's post-hoc test . P-values < 0 . 05 were considered statistically significant . | Hepatitis B virus ( HBV ) infection remains a worldwide health problem that affects more than 350 million people . HBV is one of the major etiological pathogens for liver cirrhosis and hepatocellular carcinoma . HBV covalently closed circular DNA ( cccDNA ) is a key viral intermediate for persistent infection . However , the molecular mechanism of cccDNA formation has not been clarified . Here , we found that the host factor flap-endonuclease 1 ( FEN1 ) is pivotal in cccDNA formation . We developed a novel cccDNA formation assay by the incubation of purified viral DNA with recombinant FEN1 , DNA polymerase , and DNA ligase . This study provides new insights into the molecular mechanisms of cccDNA formation and proposes FEN1 as a potential anti-HBV drug target . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"medicine",
"and",
"health",
"sciences",
"ecology",
"and",
"environmental",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"molecular",
"probe",
"techniques",
"enzymes",
"pathogens",
"gene",
"regulation",
"microbiology",
"enzymology",
"dna-binding",
"proteins",
"nucleotides",
"hepatitis",
"b",
"virus",
"viruses",
"polymerases",
"aquatic",
"environments",
"forms",
"of",
"dna",
"circular",
"dna",
"dna",
"molecular",
"biology",
"techniques",
"gel",
"electrophoresis",
"ligases",
"research",
"and",
"analysis",
"methods",
"small",
"interfering",
"rnas",
"electrophoretic",
"techniques",
"electrophoretic",
"blotting",
"proteins",
"medical",
"microbiology",
"gene",
"expression",
"microbial",
"pathogens",
"marine",
"and",
"aquatic",
"sciences",
"hepatitis",
"viruses",
"molecular",
"biology",
"biochemistry",
"rna",
"southern",
"blot",
"dna",
"polymerase",
"freshwater",
"environments",
"nucleic",
"acids",
"oligonucleotides",
"viral",
"pathogens",
"earth",
"sciences",
"genetics",
"biology",
"and",
"life",
"sciences",
"non-coding",
"rna",
"fens",
"organisms"
] | 2018 | Flap endonuclease 1 is involved in cccDNA formation in the hepatitis B virus |
Experimental studies have demonstrated that nanoparticles can affect the rate of protein self-assembly , possibly interfering with the development of protein misfolding diseases such as Alzheimer's , Parkinson's and prion disease caused by aggregation and fibril formation of amyloid-prone proteins . We employ classical molecular dynamics simulations and large-scale density functional theory calculations to investigate the effects of nanomaterials on the structure , dynamics and binding of an amyloidogenic peptide apoC-II ( 60-70 ) . We show that the binding affinity of this peptide to carbonaceous nanomaterials such as C60 , nanotubes and graphene decreases with increasing nanoparticle curvature . Strong binding is facilitated by the large contact area available for π-stacking between the aromatic residues of the peptide and the extended surfaces of graphene and the nanotube . The highly curved fullerene surface exhibits reduced efficiency for π-stacking but promotes increased peptide dynamics . We postulate that the increase in conformational dynamics of the amyloid peptide can be unfavorable for the formation of fibril competent structures . In contrast , extended fibril forming peptide conformations are promoted by the nanotube and graphene surfaces which can provide a template for fibril-growth .
The fast-developing field of nanotechnology has already had a significant impact in numerous areas of science and technology due to the ability to control the properties of nanomaterials with greater precision [1]–[3] . Despite the remarkable speed of developments in nanoscience , little is known about the effects of nanomaterials on biological matter [4] . There is a growing concern that nanomaterials , specifically those used for medical applications , may induce cytotoxic effects [5] . In addition , engineered nanomaterials , which are increasingly being used in industry and the manufacture of household goods have the ability to permeate blood-brain barriers and thus have the potential to damage cells in vivo [6] . The toxicity of nanoparticles has been associated with fibril formation , where nanoparticles can cause localization of peptides and proteins on their surfaces and promote undesirable aggregation that can favor formation of amyloid fibrils . These highly-structured protein aggregates are responsible for many degenerative diseases such as Alzheimer's , Creutzfeld-Jacob disease , and dialysis-related amyloidosis [7]–[10] . Carbonaceous nanoparticles are one of the most prevalent types of nanomaterials present in the environment . These air-borne particles are continuously injected into the atmosphere in large quantities through the process of combustion and , at the smallest scale , are in the form of clusters with nanometric dimensions . Carbon based nanomaterials , such as fullerenes , nanotubes and graphene surfaces , have been widely studied for potential applications due to their outstanding mechanical , thermal and electronic properties . There is , however , a growing volume of literature that alerts to the potential harm from both intentional ( medicinal ) and unintentional exposure of living organisms to such particles [6] , [11] , [12] . Comprehensive understanding of organic-inorganic interactions is crucial in order to minimize the potential toxicological effects associated with advances in the development and use of such nanomaterials [13] , [14] . Computational modeling has been used extensively to study the dynamic , thermodynamic and mechanical properties of biological systems . Recent reviews summarize the application of computer simulations to the study of biological matter in the presence of nanomaterials , specifically the common modes by which nanomaterials interact with proteins , DNA and lipid membranes [15]–[19] . Physicochemical properties that may be important in understanding the toxic effects of nanomaterials include particle size and size distribution , shape , exposed surface area , internal structure and surface chemistry [20] . Much research has focused on the characterization of carbon-based nanomaterials such as fullerenes , carbon nanotubes and graphene surfaces [21]–[23] . At the same time , experiments involving carbonaceous nanomaterials in biological milieu are still limited and the interactions involved are not well understood [24] . Specifically , there are some contrasting findings that have recently been published on the role of carbon nanotubes in fibril formation . Linse et al . found an increase in the rate of fibrillation by β2-microglobulin in the presence of carbon nanotubes , where they suggested that a locally increased concentration of protein on the carbon nanotubes surface promotes oligomer formation [9] . Two other separate studies also suggested that carbon nanotubes act as catalyst for fibril formation [25] , [26] . In contrast , Ghule et al . found that multi-walled carbon nanotubes inhibit amyloid aggregation of the human growth factor protein , hFGF-1 by encapsulating the protein structure and suppressing like-protein interactions [27] . Furthermore , recent computational studies of Aβ peptides found that carbon nanotubes drive the formation of β-barrels around the nanoparticle [28] , [29] . The authors suggested that this type of aggregation would lead to; 1 ) blocking of the peptide structure for further peptide association; 2 ) reducing the population of monomers/oligomers available for fibril growth; and thus resulting in an inhibition of Aβ fibrillation [28] . In addition , they proposed that the hydrophobic and π-π interactions between the Aβ peptide and carbon nanotube inhibit β-sheet formation and destabilize fibril-seeds into random coil aggregates , which would increase the nucleation lag-time and possibly reverse the fibrillation process [29] . It is evident from the works presented above that there are contrasting views on the role of carbon nanomaterials in fibril formation . However , there is substantial evidence that suggests carbon nanomaterials can have fibril inducing and inhibiting capabilities depending on the structural architecture of the nanoparticle itself and more importantly , the affinity of the peptide/protein under investigation , which plays a crucial role in the propensity for aggregation and/or fibril formation on nanoparticles [18] , [19] . While advances in experimental techniques are able to probe ever-smaller length-scales and ever-shorter timescales , atomistic modeling is a valuable complementary approach for a systematic investigation of detailed mechanisms of nanoscale phenomena at the atomistic and electronic levels [23] , [30]–[40] . Herein , we present a computational study investigating the effects of curvature and shape of carbonaceous nanomaterials on the structure , dynamics and binding of an amyloidogenic apolipoprotein C-II ( apoC-II ) derived peptide , apoC-II ( 60-70 ) . ApoC-II is a 79 amino acid protein , with an important role in lipid transport [41] , [42] . Under lipid-depleted conditions , apoC-II readily forms homogeneous fibrils with a “twisted ribbon” morphology and all of the characteristics of amyloid fibrils [43] . ApoC-II amyloid fibrils are commonly associated with atherosclerotic plaques , where they have been found to co-localize with other apolipoproteins and initiate early events in heart disease [44] . Studies have shown that airway exposure to concentrated ambient particles and single-wall carbon nanotubes can promote progression of the atherosclerosis process in apolipoprotein-E knockout mice that develop plaques in blood vessels at early age [45] , [46] . Similarly , a study by Vesterdal et al . demonstrated that intraperitoneal administration of pristine C60 fullerenes is associated with a moderate decrease in the vascular function of mice with atherosclerosis [47] . The apoC-II peptide derivative , apoC-II ( 60-70 ) , was found to have the ability to form amyloid fibrils independently [48] . This peptide has been extensively investigated under a range of conditions and in different environments using experimental and computational techniques [33] , [34] , [37] , [38] , [49] , [50] . Our previous studies using molecular dynamics simulations of the monomeric wild-type apoC-II ( 60-70 ) peptide showed that it preferentially adopts hairpin-like structures in solution . This structure was defined as an intermediate state on-pathway for the formation of fibril-seeds . Increased solvent accessible surface area and the relative orientation of the aromatic side-chains were features identified as fibril-favoring for this peptide , as they promoted hydrophobic interactions with other like-peptides . In contrast , increased flexibility and the broader distribution of angles between the aromatic residues of mutated apoC-II ( 60-70 ) resulted in slower aggregation kinetics , in other words these features were fibril-inhibiting , as demonstrated by our experiments [34] , [37] . Furthermore , our research on oligomeric apoC-II ( 60-70 ) showed that extended β-sheet structures stabilize preformed dimers and tetramers of apoC-II ( 60-70 ) . The results suggested that a tetrameric oligomer in anti-parallel configuration can serve as a possible seed for fibril formation of apoC-II ( 60-70 ) , where side-chain-side-chain contacts contribute to the fibril stability , while the maximum exposure capacity of the whole peptide ( backbone and aromatic side-chains ) promotes the growth of the fibril-seed due to the increase of exposure to other peptides [34] , [37] . Overall , the solution based studies on the behavior of apoC-II ( 60-70 ) in different environments provide benchmarking data for identifying the effects of nanomaterials on the structure and dynamics of this amyloidogenic peptide . Here , we investigate the behavior of apoC-II ( 60-70 ) in the presence of three carbonaceous nanomaterials: a spherical C60 fullerene , a tubular single-wall carbon nanotube and a flat graphene surface . We study the peptide's structure , dynamics and binding , all of which can influence its fibril formation capacity and compare the results with the previously characterized peptide behavior in solution [33] , [34] , [37] , [38] . We apply a novel combination of computational methods , including large-scale electronic structure calculations and classical all-atom molecular dynamics . This approach was recently applied for the first time to investigate the fibril inhibition mechanisms of cyclic apoC-II ( 60-70 ) and its linear analogue [39] . This combined modeling approach enables investigation of the fundamental driving forces behind the interactions of the peptide with nanomaterials and their effects on the peptide structure , dynamics and binding affinity .
To investigate the effects of carbonaceous nanomaterials on the structure and dynamics of apoC-II ( 60-70 ) ( MSTYTGIFTDQ , 169 atoms ) a series of simulations were performed with different starting peptide conformations and arrangements . The fullerene particle consisted of 60 carbon atoms with a radius of ∼3 . 5 Å . The nanotube was modeled as a ( 5 , 5 ) single-walled tube with 320 atoms in an open ended armchair arrangement , 6 . 78 Å in diameter and ∼38 Å in length , which was sufficiently long to prevent interactions between the peptide and the edges . Graphene was modeled as a periodic single sheet of 2160 carbon atoms in a hexagonal arrangement to represent an infinite graphene surface . The initial configurations were constructed by positioning the peptide 4 . 5–10 Å from the nanomaterial ( see Table S1 and S2 in Supporting Information ) . The peptide together with the nanomaterial was then placed in a periodic simulation cell of at least 60 Å×60 Å×60 Å in dimension . The molecular dynamics ( MD ) simulations were performed using the Gromacs 3 . 3 [51] simulation package , with the interactions between the particles in the system described by the united-atom Gromos forcefield and the 43A1 parameter set . The carbonaceous nanomaterials were modeled using the aromatic sp2 carbon parameters . We note that polarizable forcefields which describe the electrostatic interactions with the use of distributed multipoles [32] , [40] , [52] have been under development for graphitic structures , however , recent studies have shown that classical forcefields produce results comparable to experiment [16] , [53] , [54] . The Lennard-Jones interactions were truncated at 10 Å , with the long-range electrostatic interactions accounted for by the Particle Mesh Ewald ( PME ) method [55] . The LINCS algorithm was used to constrain the bond lengths to their equilibrium values [56] , enabling a timestep of 2 fs to be applied for all simulations . The VMD software package was used for visualization of the dynamics and analysis of the molecular trajectories [57] . In vacuo energy minimization using steepest descent algorithm was initially performed on the peptide-nanoparticle systems to remove steric clashes . The optimized system was then solvated using the SPC water model [58] at a water density of ∼1 g/cm3 . To neutralize the overall negative charge of the system , an unrestrained counterion ( Na+ ) was included in the simulation cell . Energy minimization on the solvated system was performed to relax all of the atomic degrees of freedom . Subsequently , MD was conducted to allow the solvent to equilibrate around the solutes by keeping the peptide and nanomaterial restrained . A constant pressure of 1 bar and temperature of 300 K were maintained using the Berendsen barostat and thermostat [59] . In all simulations the geometry of the nanomaterial was restrained for ease of monitoring the peptide dynamics . Two initial non-fibrillar conformations ( native and helical ) of apoC-II ( 60-70 ) peptide were simulated in the presence of each nanomaterial . To further enhance conformational sampling simulations were repeated six times with different starting orientations of apoC-II ( 60-70 ) with respect to the nanomaterial , yielding a total 800 ns of data per nanomaterial-peptide complex . The behavior and structures observed in each system exhibited distinctive trends , therefore the results from representative simulations are shown . Using umbrella sampling together with the weighted histogram analysis method ( WHAM ) [60] , potential of mean force ( PMF ) profiles were generated to evaluate the free energy of dissociation ( ΔG ) for apoC-II ( 60-70 ) bound to each nanomaterial in solution . This method was applied to explicitly solvated systems and therefore accounts for the entropic contributions in the determination of the dissociation energies . We determined the PMF as a function of separation distance between the center of mass of the nanomaterial and the α-carbon of the glycine residue in apoC-II ( 60-70 ) . To acquire the PMF profiles , a series of simulations ( windows ) were performed at increasing distance between the peptide and nanomaterial , starting from typical equilibrium structures of the peptide-nanomaterial complex . The peptide was restrained at each window using Hookean functions with a force constant of 8000 kJmol−1 nm−2 . In the present work , ΔG and PMF both refer to the free energy required to bring the peptide and nanomaterial from an associated form , which defines our zero of free energy , to some separation d . Adjacent windows were separated by 0 . 5 Å and each window was simulated for 15 ns with at least 30 windows used ( until the peptide was fully dissociated from the nanomaterial ) , resulting in a total simulation time of at least 450 ns per nanomaterial complex . WHAM was subsequently applied on the final 5 ns of simulations to remove the biasing potentials and obtain the unbiased PMF profiles . The overlap between neighboring windows was monitored to ensure the suitability of the selected spring constant and sufficient conformational sampling ( not shown ) . Binding energy calculations were performed on multiple structures selected from the classical forcefield simulations of each peptide-nanomaterial system ( more details in aromatic tracking section ) . Based on the findings of a recent methodological study , classical energy minimization was performed in solution prior to electronic structure calculations [61] . This procedure reduced any electrostatic artefacts that may arise due to the electronic structure calculations being performed in vacuum while retaining the major structural features of the system obtained during the fully solvated MD simulations . Single point electronic energy calculations performed on the resultant frames were used to calculate in vacuo binding energies between the peptide and nanomaterial . We determined the binding energy ( Eb ) of apoC-II ( 60-70 ) peptide on each nanomaterial , as ( 1 ) where EP+N is the total energy for the peptide-nanomaterial complex , EP is the total energy of the apoC-II ( 60-70 ) peptide and EN is the total energy of the isolated nanomaterial . The linear-scaling DFT code ONETEP [62] was used , which combines linear scaling computational efficiency with accuracy that is comparable to traditional plane-wave DFT codes . Such efficiency opens up the possibility of performing accurate DFT calculations on thousands and tens of thousands of atoms , including proteins [63]–[65] and various nanomaterials [66] , [67] . ONETEP achieves linear scaling by exploiting the ‘near-sightedness’ of the single-particle density matrix p ( r , r' ) in non-metallic systems , ( 2 ) where K is the density kernel and φα are a set of strictly localized non-orthogonal generalized Wannier functions ( NGWFs ) [68] . The total energy is self-consistently minimized with respect to both the density kernel and the NGWFs . The NGWFs are expanded in a basis set of periodic sinc ( psinc ) functions [69] , which are equivalent to a plane-wave basis , and are optimized in situ , giving plane-wave accuracy and allowing the accuracy to be systematically improved with a single kinetic energy cut-off parameter . The PBE generalized-gradient approximation was used to describe exchange and correlation [70] , and norm-conserving pseudopotentials were employed to describe the interactions between electrons and nuclei . Dispersion interactions were accounted for using a DFT+D approach [71] . Dispersion-corrected DFT has been shown to produce accurate results for weakly interacting systems , such as aromatic composites [72] and protein-ligand complexes [73] . The supercell dimensions for each system were sufficiently large to prevent interactions between periodic images . In all cases , NGWF radii of 8 bohr were used for all atoms , no truncation was applied to the density kernel , the kinetic energy cut-off for the psinc basis was 880 eV , and the Brillouin zone was sampled at the Γ-point only . The role of aromatic residues in the adsorption of apoC-II ( 60-70 ) to each nanomaterial was investigated by tracking the placement of the peptide's aromatic rings across each graphitic surface . This technique determines the position and orientation of the aromatic rings in amino acids relative to the rings within the nanomaterials' surface at every step of the MD trajectories . The aromatic ring arrangement was categorized into three groups: no π-stacking , offset π-stacking and face-to-face π-stacking ( Figure 1 ) . The criteria to determine no π-stack register were a pair-wise contact distance over 4 . 5 Å between any two atoms of the aromatic ring and nanomaterial; or an angle greater than 30° between the plane normal of the aromatic ring and nanomaterial surface [74] . Face-to-face π-stacking was accounted for when the displacement between the centroids of the phenyl rings of the aromatic residues and centroid of the nearest hexagonal carbon ring was less than 0 . 71 Å ( half the carbon-carbon bond length ) . A displacement greater than 0 . 71 Å was considered as offset π-stacking . ApoC-II ( 60-70 ) peptide having two aromatic rings in its sequence ( Tyr63 and Phe67 ) resulted in six possible ring arrangements relative to the nanomaterials surface . The categories were defined as: ( 1 ) no π-stacking by both rings; ( 2 ) offset π-stacking by one ring and no π-stacking by the other; ( 3 ) offset π-stacking by both rings; ( 4 ) face-to-face π-stacking by one ring and no π-stacking by the other; ( 5 ) face-to-face π-stacking by one ring and offset π-stacking by the other; ( 6 ) face-to-face π-stacking by both rings . Once the aromatic arrangement was categorized , each group underwent structural clustering with RMSD cut-off of 2 Å for the entire peptide using the single linkage clustering method to determine the most frequently sampled structure within each π-stacking category . Three representative structures from each π-stacking group were selected and underwent electronic structure calculations to determine their binding energies .
Secondary structure analysis was performed to investigate the effects of the nanomaterial curvature on the peptide's conformation . The STRuctural IDEntification ( STRIDE ) [53] algorithm was utilized to classify the peptide's secondary structure as a function of time . Secondary structure evolution plots depicting typical conformational trends exhibited by apoC-II ( 60-70 ) in the presence of C60 , nanotube and graphene are shown in Figure 2 . The initial 20 ns of simulation ( equilibration ) are also shown to highlight the conformational changes in the peptide induced by adsorption onto the nanomaterial surface . The results show a structural transformation of the peptide upon adsorption to the nanomaterial surface . ApoC-II ( 60-70 ) in the presence of C60 was observed to curve around the particle , with a turn region around Gly65 , as shown in the picture inset of Figure 2a . This structure allows for a large number of contacts to be made with the nanoparticle , dominated by π-interactions between the aromatic residues ( Tyr63 and Phe67 ) and the C60 surface . Due to the presence of C60 , apoC-II ( 60-70 ) is unable to form the inherent β-hairpin conformation [34] , [37] . In one out of six simulations the peptide was able to dissociate within 10 ns of contact with the C60 particle . This suggests that the peptide can be weakly bound to the surface of the nanoparticle . Upon desorption apoC-II ( 60-70 ) was able to form a β-hairpin conformation ( picture inset of Figure 2b ) . We note that the β-hairpin structure was found favorable for monomeric apoC-II ( 60-70 ) peptide in solution , and identified as an intermediate state on-pathway for fibril formation [34] , [37] , [75] . The results show that the presence of C60 inhibits the formation of the characteristic fibril favoring β-hairpin as well as the extended conformation suggesting that the interactions with the C60 may contribute to an increase in mobility ( see Figure S1 ) and facilitate the formation of fibril incompetent conformations . Recent work by Andujar et al . where they showed that C60 induced significant destabilization of the amyloid-β fibrils by disrupting the hydrophobic contacts and salt-bridges between the β-sheets [76] is in line with our work . This suggests that C60 can be used as a prototype for the design of potential fibril inhibitors . The secondary structure evolution plot of apoC-II ( 60-70 ) in the presence of a nanotube shows that the peptide exhibits different structural features compared to those in the presence of C60 . The peptide tends to elongate across the surface of the nanotube , while adopting mostly turn and coil motifs . This behavior is a result of the large surface area available for contact on the nanotube ( Figure 2c ) . The strong affinity between the nanotube and peptide is assisted by π-π interactions between the aromatic rings of the peptide and the nanotube . The curvature of the nanotube enables the peptide to arch , which facilitates the short-lived formation of a hydrogen bond between Ser61 and Thr64 . In comparison to the C60 simulations , the peptide was less dynamic on the surface of the nanotube , as seen from the smaller number of conformations sampled by the peptide following the adsorption and immobilization on the nanotube surface ( Figure 2c and Figure S1 ) . The simulations of apoC-II ( 60-70 ) in the presence of graphene exhibited similar structural features to those seen in the presence of the nanotube . Upon adsorption , the peptide elongates along the graphene surface and features predominantly turn and coil structures ( Figure 2d ) . The large surface area available for interactions enables the peptide to freely slide on the surface , while the favorable π-π stacking interactions between the aromatic residues of the peptide and the surface define its conformational features . Linse et al . showed extended nanoparticles enhance the probability of appearance of a critical nucleus for nucleation of protein fibrils , albeit for a different combination of nanomaterials and peptides [9] . This feature was determined as fibril-favoring in our previous works on apoC-II ( 60-70 ) oligomers [34] , [37] . Other studies have also shown that carbon nanotubes and graphene surfaces facilitate a change in the conformation of peptides [30] , [77] and π-stacking is an efficient mode of biological recognition of π-electron-rich carbon nanoparticles [23] , [30]–[32] , [52] , [77] , [78] . A common feature in all secondary structure plots is the presence of a persistent coil motif at the C-terminal end of the peptide , where predominantly hydrophilic residues reside . This suggests that the inherent preference for interaction with the polar environment by these residues is suppressed by the attractive van der Waals forces between the large surfaces presented by the nanomaterial and the peptide , preventing the peptide dissociation from the nanomaterial . Strong hydrophobic interactions between the WW domains and carbon nanotubes have also been associated with protein function “poisoning” and disruption of the protein active site [79] . Overall , it should be noted that the adsorbed peptide may adopt both fibril initiating as well as fibril incompetent conformations . However , our analyses indicate that the extended conformation adopted on the extended nanosurfaces is in line with the fibril competent structures we found through our previous modeling and experimental studies [33] , [34] , [37] , [38] , [49] , [50] . In contrast the mobility and lack of secondary structure elements needed for the fibril formation by the C60 adsorbed peptide suggests the inhibiting role of this nanoparticle in the fibril formation . To gain a more detailed understanding of the interactions involved in adsorption of apoC-II ( 60-70 ) , the contact stabilities of the peptide's residues with each nanomaterial were investigated ( Figure 3 ) . Contact stabilities were calculated as the percentage of simulation time during which a contact was maintained between each residue and the respective nanomaterial . A contact was counted when the distance between a pair of atoms was less than 4 Å , which enabled us to account for van der Waals interactions between the peptide and the nanoparticle . High contact stabilities were found for both the aromatic tyrosine ( Tyr63 ) and phenylalanine ( Phe67 ) residues with all nanomaterials . The stable π-stacking arrangements between the aromatic rings of the peptide and the electron-rich carbon rings of the surface , suggest that these are the key residues that contribute to the strong interactions between the peptide and the carbonaceous nanomaterials , in line with other studies [23] , [30]–[32] , [52] , [77] , [78] . This effect is evident in all simulations , however in the C60 complex the aromatic residues dominate the interactions between the peptide and C60 surface , while the other residues exhibit less persistent contacts . Interestingly , Tyr63 exhibits higher binding affinity to C60 compared to Phe67 , in accordance with a DFT study that showed Phe and Tyr bind with a similar strength to the nanotube , while Tyr exhibits a stronger binding to C60 [80] , [81] . In contrast , the large contact area presented by the carbon nanotubes and graphene results in higher contact stabilities with all ( not just the aromatic ) residues , this effect being most evident on the graphene surface . In our recent studies we showed that the orientation of the aromatic side chains is different in the fibril-forming and fibril-inhibiting arrangements [34] , [37] . The simulations of apoC-II ( 60-70 ) in the presence of C60 exhibited structures where the aromatic rings were positioned on the same side of the peptide which enhance the π-stacking interactions with the small , highly curved C60 particle . This ring arrangement was postulated to inhibit fibril formation [34] , [37] . In contrast , the aromatic rings did not show a specific facial preference in the nanotube and graphene complex simulations ( Figure 2c , d ) . This was due to the large contact area and stronger hydrophobic interactions presented by these materials , which formed the aromatic ring stacking upon adsorption of the peptide . A series of radial distribution functions ( RDF ) were calculated to determine the degree of water structuring around the peptide in solution and when bound to the nanomaterial surface . The results provide an insight into the extent of desolvation of the peptide conformation upon binding to the different nanomaterials . Typical RDFs of the peptide side-chain hydrogen atoms ( H ) with respect to the water oxygen atoms ( O ) are shown in Figure 4 . For all systems , the RDF profiles show a peak at ∼2 Å , representing the first hydration shell , indicating hydrogen bonding between water and the apoC-II ( 60-70 ) side-chains . The results also show the presence of a second hydration shell at ∼4 Å . The RDFs of peptide-nanomaterial complexes exhibit an attenuation of the overall probability density , suggesting the exclusion of water due to the hydrophobic contact between the peptide and nanoparticle manifests in the RDFs through the lowering of the occurrence of water at larger separation distances in the bound state . Indeed , desolvation effects have been shown to be favorable in the self-assembly of cyclic peptides on carbon nanotubes [31] . To characterize the binding of apoC-II ( 60-70 ) peptide to each nanomaterial in the presence of solvent , the free energy of dissociation was calculated using umbrella sampling ( potential of mean force , PMF ) together with the weighted histogram analysis method ( WHAM ) [60] . This approach is applied to explicitly solvated systems and accounts for both the enthalpic and entropic contributions to the dissociation free energies . Two bound equilibrium complex structures were studied for each system to enhance sampling . The PMFs detailing the dissociation pathway of apoC-II ( 60-70 ) from each nanomaterial are presented in Figure 5 . The results demonstrate that higher degree of curvature reduces the surface area available for adsorption , and the dissociation free energy indicates that binding to C60 is weakest and binding to graphene is strongest of the systems investigated . Here , a lower value indicates a weaker binding . The size of the peptide does not allow for complete wrapping of the C60 , therefore in this complex the peptide is quite mobile with terminal residues remaining free and not forming close contacts with the nanoparticle , as can be seen in the residue contact stability plot in Figure 3 . Figure 5a indicates that the dissociation energy is dependent on the adsorbed peptide conformation . This suggests that C60 induces significant structural lability in apoC-II ( 60-70 ) preventing it from adopting stable conformations , in line with the peptide evolution observed through molecular dynamics trajectories ( Figure 2 and Figure S1 ) . A higher free energy of dissociation ( ∼1 . 8 kcal/mol ) was obtained for the peptide that had a larger number of contacts with C60 and whose aromatic rings were continuously interacting with the C60 particle . In contrast , the system where the two aromatic rings of the peptide predominantly formed π-stacking between themselves rather than with the nanoparticle resulted in a lower dissociation energy ( ∼1 . 1 kcal/mol ) . The peaks and troughs are caused mostly by the transient π-stacking interactions , with peaks observed when contacts are broken , as illustrated by the insets in Figure 5 . In our previous work on apoC-II ( 60-70 ) we showed that an increase in conformational flexibility and dynamics can slow down or even inhibit fibril formation [34] , therefore it appears that interactions with the C60 can induce a similar , fibril inhibiting , effect . We note that generally PMF plots for the C60-peptide system are noisier than those for the nanotube and graphene systems which is due to the transient nature of the contacts and increased mobility of apoC-II ( 60-70 ) when in contact with C60 , rather than due to insufficient conformational sampling . The effect was verified by continuing the umbrella sampling simulations for a further 15 ns per window and observing that the resultant PMFs did not show significant differences ( figures not shown ) . Similarly , the free energy differences seen between the multiple simulations of the peptide-nanotube system are due to the variety of structures sampled along each dissociation pathway . As expected , the predominantly elongated peptide conformation ( Figure 5b , left ) enabled a larger number of contacts between the peptide and the nanotube , which resulted in a higher free energy of dissociation ( ∼8 . 6 kcal/mol ) . The peptide exhibiting mostly coiled structures made fewer contacts with the nanotube ( Figure 5b , right ) which in turn required less energy ( ∼3 . 5 kcal/mol ) to dissociate from it . The smoother PMF plots for the peptide-nanotube systems are a result of the persistent interactions between the components , in keeping with the results of our classical MD simulations . The PMF plots representing the dissociation free energy of apoC-II ( 60-70 ) from graphene exhibited conformation independent pathways . As seen from the MD results , the π-stacking between apoC-II ( 60-70 ) and graphene contributes to the formation of elongated peptide structures and restricts the conformational flexibility of the peptide . Repeat simulations resulted in dissociation free energies of ∼15 kcal/mol irrespective of the conformations sampled along the dissociation pathway . We note that dissociation energy peaks occur when a large number of contacts are broken , such as during the illustrated dislocation of the aromatic residues from the graphene surface ( see insets of Figure 5c ) . In contrast , as the peptide is slowly pulled away from the surface a characteristic smooth dissociation energy profile is observed . In addition to the classical simulation-derived dissociation free energies discussed above , we have used electronic structure calculations based on DFT to calculate in vacuo binding energies of selected frames derived from classical all-atom simulations . To investigate the role of aromatic residues ( Tyr63 and Phe67 ) in driving the adsorption of apoC-II ( 60-70 ) onto carbon based nanomaterials , we developed an algorithm capable of tracking the position and orientation of the phenyl rings of the aromatic amino acids with respect to the aromatic rings of the nanomaterials' surface at every step of the MD trajectories ( exemplar result shown in Supporting Information ) . The in vacuo binding energy of three representative frames from each π-stacking arrangement was obtained by DFT calculations using the ONETEP linear-scaling code [62] , comprising a total of eighteen typical structures per peptide-nanoparticle complex . This analysis provides a measure of the relative binding of apoC-II ( 60-70 ) to a nanomaterial surface with respect to the contact area . The binding energy differences between the representative structures for each π-stacking configuration versus the peptide-nanomaterial contact area are shown in Figure 6 ( tabulated form available in Supporting Information ) . The vacuum binding energies are shown relatively to the strongest bound state ( apoC-II ( 60-70 ) on graphene ) . In this case , higher values indicate a weaker binding ( left y-axis , Figure 6 ) . The in vacuo binding energy results confirm the trends observed in our explicitly solvated PMF free energies showing the strength of binding between apoC-II ( 60-70 ) and the nanomaterials to follow: C60<nanotube<graphene . In all systems the aromatic rings act like “anchors“ for binding the peptide to the carbon nanomaterials via π-π interactions . DFT binding energy calculations confirm the finding from classical MD that apoC-II ( 60-70 ) exhibits strongest binding on graphene with a face-to-face π-stacking arrangement made by the two aromatic rings of the peptide and the surface . The all-atom MD simulations show that the flat graphene surface promoted sliding of the peptide ( see Figure S2 in Supporting Information ) and backbone elongation to optimize the π-stacking arrangement between the aromatic rings of the peptide and the substrate . This contributes to the peptide-graphene system having the largest aromatic and total contact area , which results in the strongest binding . The DFT binding energy also confirmed that apoC-II ( 60-70 ) exhibits a weaker binding to the nanotube and the weakest binding to C60 , attributed to the increased nanosurface curvature which stimulates the formation of turns and loops in apoC-II ( 60-70 ) leading to a lower contact area between the peptide and nanoparticle . We note that C60 comprises both hexagonal and pentagonal carbon rings and , therefore , has a lower probability of face-to-face π-stacking with the six-membered aromatic rings of the peptide ( statistical data shown in Supporting Information ) . This provides a further explanation for the significantly smaller contact area and weaker binding obtained for the peptide and C60 nanoparticle , compared to the nanotube and graphene systems ( Figure 6 and Table S3 ) . Furthermore , using our DFT calculations we were able to examine the intra-peptide electrostatic interactions which play a significant role in determining the peptide's secondary structure and consequently the binding affinity to other materials . Electron density difference ( Δρ ) maps showing charge accumulation ( red ) and depletion ( blue ) upon peptide adsorption on each nanomaterial are presented in Figure 7 . We can see that intra-peptide interactions are more significant in the proximity of nanoparticles with high curvature which have a reduced nanoparticle-peptide contact surface area , as Figures 6 and 7 demonstrate . The greater surface area available on the flatter “hexagonal-only” surface of graphene allows for a more efficient π-stacking and a stronger peptide binding as shown in Figures 7c . Moreover , surface adsorbed elongated peptide conformations enable polar residues such as Thr , Ser and Gln to become more solvent exposed , thus exhibiting the “snorkeling effect” [82] , [83] ( see inset of Figure 7b ) , where the hydrophobic backbone interacts with the graphitic surface , while the polar side chains are protruding to the solvent . This lowers the overall contact area between the peptide and nanomaterial , and ultimately reduces the binding affinity . Figure 7 shows small electron density differences between the aromatic groups and the graphitic surfaces . This is in agreement with the study of Poenitzsch et al . where they observed weak charge-transfer interactions between aromatic groups and carbon nanotubes using scanning tunneling spectroscopy and Raman experiments [84] . Our electron density analysis shows that , generally , a weaker binding is a result of inefficient π-stacking arrangements and intra-peptide electrostatic interactions that reduce the peptide-surface interactions , as Table S3 demonstrates . Charge redistribution can also be seen between the peptide and nanoparticle surface ( Figure 7 ) , suggesting some polarizability effects occur between the peptide and nanomaterial . Specifically , a charge depletion can be seen at Asp69 and Gln70 in all systems , while a charge accumulation develops at the closely interacting sites of the peptide and nanomaterial surface . Figure 7a shows charge accumulation at Gly65 for the C60 complex , while the nanotube and graphene exhibit charge buildup in close proximity to the predominantly hydrophilic N-terminal region of the peptide ( Met60 , Ser61 and Thr62 ) . We note that the agreement between our classical simulations and the DFT studies suggest that the classical forcefield potentials employed here are able to capture the polarization effects inherent to peptide-nanoparticle systems . Moreover , a recent study using the dispersion corrected DFTB-D method , showed that although molecular mechanics techniques with fixed-charge forcefields do not explicitly incorporate polarizability , they can predict the strength of π-π interactions between aromatic moieties and carbon nanotubes [75] . This demonstrates that molecular dynamics simulations utilizing fixed charge forcefields provide a reasonable representation of the interactions between peptides and graphitic surfaces . Using classical forcefield and electronic structure calculations , we have shown that an amyloidogenic apoC-II ( 60-70 ) peptide exhibits a strong affinity for graphitic nanomaterials where binding is facilitated through π-π interactions between the aromatic residues of the peptide and the surface of the nanomaterial . This is generally achieved by the exclusion of water molecules from the peptide-nanomaterial interface . The proximity of the C60 fullerene contributed to an increase in conformational lability of apoC-II ( 60-70 ) , which was shown to prevent it from adopting fibril-favoring structural features . This finding is in line with the previous studies of oxidized apoC-II ( 60-70 ) , where increased structural flexibility and dynamics were the key factors prohibiting this peptide to form fibrils , confirmed experimentally . Conversely , our data showed that the less curved nanotube and flat graphene nanomaterials promote elongated peptide conformations previously shown to form fibril seeds , which confirms recent findings that extended carbon nanosurfaces can act as templates able to encourage peptide fibril formation and growth . Electronic binding energy and solution free energy calculations showed the binding affinity of apoC-II ( 60-70 ) was weakest for the C60 particle , followed by the nanotube , and strongest for the graphene . In all simulations these trends are due to the larger contact area available for peptide adsorption to the flatter graphene and nanotube than the highly curved C60 . The increased curvature also results in reduced efficiency of aromatic π-stacking and higher intra-peptide electrostatic interactions which contributes to its weaker binding to the nanomaterials . The electronic structure calculations show that dimensionality that determines the electronic properties of the nanoparticle as well as size and curvature play a significant role in the contact area and binding mechanisms of the peptide . At the same time the intra-peptide interactions determined by the peptide sequence ( i . e . presence of aromatic , aliphatic , polar/apolar amino acids ) affect the binding mechanism of peptides to nanoparticles . The observed agreement between the classical and electronic structure calculations show that molecular dynamics simulations utilizing fixed charge forcefields provide reasonable representation of the interactions between peptides and graphitic surfaces . In summary , our results highlight that hydrophobic nanoparticles have multiple notable effects on the peptide structure , dynamics and binding affinity . We have demonstrated that dimensionality and different degree of curvature can either facilitate or hinder the interaction of amyloidogenic peptides with the nanosurfaces and make them adopt conformations capable of inhibiting or promoting fibril development , as shown in our recent experiments . These findings can be important for rational design of amyloid fibril inhibitors as well as for clarification of possible toxic effects of carbon based nanomaterials . | Investigation of the effects of nanomaterials on biological systems is crucial due to the increasing exposure to nanostructured materials with the growing developments and applications of nanotechnology in everyday life . Nanoparticles have been shown to have an effect on protein structure and interfere with protein self-assembly leading to the development of amyloid fibrils responsible for many debilitating diseases , such as Alzheimer's , Parkinson's and prion related diseases . Computational techniques enable investigation of such systems at the atomistic and electronic levels providing insight into properties not available from experiments . We employ a novel combination of computational methods , including large-scale electronic structure calculations and classical molecular dynamics to investigate the behavior of amyloidogenic apoC-II peptide in the presence of carbonaceous nanoparticles , the most prevalent form of nanoparticles found in the environment . Our results showed that carbon nanoparticles have significant effects on the peptide structure , dynamics and binding affinity . Specifically , the dimensionality and curvature of the nanomaterial can either facilitate or hinder their interaction with amyloidogenic peptides and make them adopt conformations capable of inhibiting or promoting fibril growth . These findings are important for rational design of amyloid fibril inhibitors as well as the elucidation of possible toxic effects of carbon based nanomaterials . | [
"Abstract",
"Introduction",
"Methods",
"Results/Discussion"
] | [] | 2013 | Dimensionality of Carbon Nanomaterials Determines the Binding and Dynamics of Amyloidogenic Peptides: Multiscale Theoretical Simulations |
Hereditary spastic paraplegias ( HSPs ) comprise a group of genetically heterogeneous neurodegenerative disorders characterized by spastic weakness of the lower extremities . We have generated a Drosophila model for HSP type 10 ( SPG10 ) , caused by mutations in KIF5A . KIF5A encodes the heavy chain of kinesin-1 , a neuronal microtubule motor . Our results imply that SPG10 is not caused by haploinsufficiency but by the loss of endogenous kinesin-1 function due to a selective dominant-negative action of mutant KIF5A on kinesin-1 complexes . We have not found any evidence for an additional , more generalized toxicity of mutant Kinesin heavy chain ( Khc ) or the affected kinesin-1 complexes . Ectopic expression of Drosophila Khc carrying a human SPG10-associated mutation ( N256S ) is sufficient to disturb axonal transport and to induce motoneuron disease in Drosophila . Neurofilaments , which have been recently implicated in SPG10 disease manifestation , are absent in arthropods . Impairments in the transport of kinesin-1 cargos different from neurofilaments are thus sufficient to cause HSP–like pathological changes such as axonal swellings , altered structure and function of synapses , behavioral deficits , and increased mortality .
Hereditary spastic paraplegia ( HSP ) is a group of genetically heterogeneous neurodegenerative disorders characterized by distal axonopathy that affects the longest axons in the corticospinal tract [1] , [2] . To date , 48 HSP loci have been described . The three most common causes of HSP - accounting for more than 50% of all cases - are mutations in SPG3A ( Atlastin ) , SPG4 ( Spastin ) and SPG31 ( Reep1 ) . Both Atlastin and Spastin mutations as well as mutations in 6 other identified SPG genes: ( KIF5A , Nipa , Spatacsin , Spastizin , Spartin and Maspardin ) have been implicated in disturbances of the intracellular transport . This suggests that perturbations in long-range , tubulin based transport might be a common pathological mechanism underlying different forms of HSP ( for review see [3] ) . SPG10 is inherited in an autosomal-dominant manner , with age of onset varying from childhood to the fourth decade of life [4] . The SPG10 gene KIF5A encodes the heavy chain of the neuronal microtubule motor kinesin-1 [5] , [6] . The kinesin-1 family is the major anterograde motor complex . To date , 21 different SPG10 mutations have been described [5] , [6] , 19 of which localize to the motor domain of KIF5A . Neither genomic deletions nor “truncating” mutations were identified as causes of autosomal-dominant SPG10 , suggesting that SPG10 may be caused by a dominant-negative effect rather than by haploinsufficiency . We generated a Drosophila model for SPG10 to validate the proposed dominant-negative interaction between mutant and wild-type Khc in the context of a living organism . Our results imply that SPG10 is not caused by haploinsufficiency but by the loss of endogenous kinesin-1 function . Thereby , KhcN262S acts as an antimorph , not as a neomorph .
In a previous in vitro study , four point mutations in the KIF5A gene that cause HSP in humans were analyzed [7] . One of these mutations , N256S , caused a reduction of motor velocity and displayed a dominant-negative effect on the function of wild-type KIF5A at physiologically relevant ratios of mutated and wild-type kinesin , but did not influence its microtubule binding affinity [7] . The N256S mutation was selected to generate the first in vivo Drosophila model for SPG10 . Using this model we aimed at demonstrating that the mutated protein is stable in the context of an intact organism . Consistent with in vitro results [7] human genetic studies suggest [5] , [6] that SPG10 is not caused by haploinsufficiency but by the dominant-negative interaction of mutated and wild-type kinesin . Thus , we wanted to prove that the dominant-negative action of mutated kinesin persists in vivo at physiologically relevant ratios , and is not abolished by cellular quality control mechanism , which might either prevent the hetero-dimerization of mutant and wild-type kinesin or which might selectively destabilize these dimers . The KIF5AN256S mutation , which corresponds to the amino acid exchange N262S in Drosophila kinesin heavy chain ( khc ) ( Figure 1A , red star ) , is located on loop 11 . Loop 11 ( Figure 1A , red line; Figure 1B , brown cylinder ) connects the microtubule- ( α-helix 4: Figure 1A , green box; Figure 1B , yellow helix ) and the ATP-binding site ( β-sheet 7: Figure 1A , blue arrow; Figure 1B , pink β-sheet 7 ) of Khc [8] . Five of the 21 described human mutations in KIF5A ( Figure 1A , red dots ) map to seven amino acids ( EAKNINK ) of loop 11 [9] , highlighting the importance of this structure , which is essentially 100% conserved between nematodes , arthropods , and chordates ( Figure 1A ) . We used two complementary approaches to address the putative dominant negative action of KhcN262S in Drosophila . Both involve - unlike previous studies of loss of function alleles [10]–[16] - the ectopic expression of Khc in the wild-type background . Therefore , pathological alterations should only occur if KhcN262-Khcwt heterodimers are dysfunctional or if dimers containing KhcN262S are directly toxic . Ectopic expression of KhcN262S was induced either alone or in combination with Khcwt . The phenotypic severity of an antimorphic mutation will be decreased by increasing wild-type gene dosage . If KhcN262S acts as an antimorph , coexpression of Khcwt should ameliorate all observed defects . Strong expression of KhcN262S in motoneurons ( D42-Gal4 ) of wild-type larvae raised at 25°C or 29°C caused severe pathological symptoms , resulting in the death of larvae in the L2 or L3 stage , respectively . Initially , larvae display the characteristic tail-flipping phenotype originally described for khc null mutant larvae [13] . As the paralysis progresses , it ascends from posterior to anterior until larvae can only move their head ( Figure 1C , red arrow; Video S1 ) . Complete paralysis and death ensue . This “ascending” paralysis mirrors human pathological symptoms characterized predominantly by affliction of the lower limbs due to the particular vulnerability of the long descending spinal tracts . In Drosophila , axons innervating the posterior segments are considerably longer than those innervating anterior segments and seem to be - just as in humans - primarily affected . In larvae coexpressing both KhcN262S and Khcwt ( Khcwt+N262S ) , phenotypes are ameliorated , but still persist . This suggests that KhcN262S acts as an antimorph , not as a neomorph . At the age at which KhcN262S-expressing larvae are only able to move their head , Khcwt+N262S larvae can still locomote while displaying a tail-flipping phenotype ( Figure 1C , green arrowhead; Video S2 ) . This delayed paralysis of Khcwt+N262S larvae is progressive but does not cause larval lethality ( Figure 1C ) . Khcwt overexpression alone does not cause any obvious defects at either 25°C ( Figure 1C; Video S3 ) or 29°C . When the flies were raised at 18°C , decreased expression levels of KhcN262S or Khcwt+N262S resulted in extremely short-lived flies or flies with a 25% reduction in life span ( Figure 1D ) . Expression of Khcwt led to only a minimal but significant reduction in life span ( Figure 1D , 1E ) . Furthermore , the behavior of flies expressing KhcN262S at 18°C was strongly impaired . If forced to fly , animals could not sustain stable flight , but fell to the ground . At rest , they held their wings in an abnormal position ( Figure 2A ) . Whereas control flies ( Figure 2A , green arrowhead ) held their wings parallel to their body axis , KhcN262S-expressing flies held their wings up ( Figure 2A , red arrow ) . This behavior has been previously described in Parkinson disease-related Drosophila pink-1 mutants , in which the wing posture defect was attributed to apoptotic degeneration of the indirect flight muscles [17] . However , we did not observe signs of muscle loss in KhcN262S-expressing flies when thoracic indirect flight muscles were analyzed ( Figure 2B ) , suggesting that the abnormal wing posture is secondary to impaired motoneuron function and not a result of muscle degeneration . In the above-described models , KhcN262S is expressed throughout development . To test whether conditional expression of KhcN262S ( elav-Gal4 , tub-Gal80ts ) after completion of development is sufficient to cause neurodegeneration , we induced expression 0 to 24 hours after eclosion . Thirteen days of overexpression at 29°C led to a 3–4 fold excess of ectopic Khc compared with endogenous Khc ( Figure 2C ) . Importantly , equal protein levels of Khcwt and KhcN262S were detected , showing that N262S does not affect protein stability ( Figure 2C ) . Sixteen days after initiation of conditional overexpression , flies overexpressing KhcN262S were essentially unable to climb a vertical plastic surface ( Figure 2D ) , whereas flies overexpressing Khcwt showed no obvious locomotion defects . In summary , KhcN262S acts as an antimorph , not as a neomorph . Either chronic or conditional expression of KhcN262S is sufficient to cause HSP-like pathological changes in Drosophila . To further validate that KhcN262S interferes with the function of wild-type Khc in a dominant-negative manner , we compared the effects of ectopic expression of KhcN262S in a wild-type background to those observed in heterozygous and homozygous khc mutants . To this aim four to five day old larvae were selected . The expressions of KhcN262S lead - likely due to a decreased overall fitness of these larvae - to delays in larval growth as exemplified by the decreased size of 120 h old larvae ( D42>w1118: 3 . 3 mm , D42>khcN262S: 2 . 4 mm , p<0 . 001 , Student's T-Test , two-tailed ) . D42>khcN262S larvae showed also a trend to transit from the L2 to L3 stage being 15% smaller ( data not shown ) . We thus decided to analyze the locomotion speed of larvae of the same size rather than of the same larval stage . This is reasonable as the length of a larva is the main biophysical parameter promoting or limiting the fast movement of a larva . The use of Animaltracer - a custom build algorithm - allowed us to determine both speed and length of larvae in a non-biased , automated manner . We next quantified the average speed of small ( 1–3 mm ) and large ( 3–5 mm ) larvae ( Figure 3A ) . Locomotion of KhcN262S-overexpressing and khc deficient larvae was dramatically impaired compared with control and Khcwt-overexpressing larvae . Whereas locomotion of small Khcwt+N262S-expressing larvae was indistinguishable from that of controls , large larvae displayed a significant reduction in locomotion speed . No difference in the locomotion speed between wild-type larvae and larvae lacking one copy of khc were reported . These data support the hypothesis that SPG10 is caused by a dominant-negative action of mutated KIF5A that dramatically reduces endogenous kinesin-1 function . Consistent with the degenerative nature of HSP , locomotion defects caused by the ectopic expression of mutated Khc were generally more pronounced in larger larvae . In the khc null mutant , the accumulation of the synaptic vesicle ( SV ) protein , cysteine string protein ( CSP ) , has been reported [13] . We sought to address whether expression of KhcN262S and of Khcwt+N262S is sufficient to cause the accumulation of synaptic cargos in nerves . CSP has been reported to be a cargo of Khc [13] , [18] . The expression of KhcN262S and of Khcwt+N262S led to the accumulation of cargos in axons ( Figure 3B ) . Both the number and area fraction of nerves that were positive for cargo accumulations were highest in khc mutant larvae and in larvae overexpressing KhcN262S ( Figure 3C ) . Axonal cargo accumulations were absent in Khcwt-expressing larvae . Similar results were obtained by staining for another SV protein: the Drosophila vesicular glutamate transporter ( DVGlut ) ( Figure 3D , 3E ) [19] , [20] . Accumulations of cargo coincided with an increased intensity of the neuronal membrane marker anti-HRP ( arrowheads in Figure 3B , 3D , 3F , 3H ) . Next , we scored for nonselective disturbances of axonal transport . Thereby we could show that both the transport of dense core vesicles ( Figure 3F , 3G; visualized by ANF-GFP ) , [21] and the transport of the active zone protein Bruchpilot ( Brp ) [22] , [23] is disturbed in KhcN262S expressing larvae ( Figure 3H , 3I ) . Both cargos are transported by the kinesin-3 family member unc-104 [18] , [24]; indicating that both the fast axonal transport of kinesin-1 and kinesin-3 cargos is disturbed by the expression of mutated Khc . We then aimed to validate that accumulations of cargo characterized by strong fluorescence for both anti-HRP and SV ( Figure 3B , 3D , arrowhead ) are axonal swellings . A swelling is defined by ( 1 ) the accumulation of cargo , ( 2 ) a local increase in anti-HRP staining intensity , and ( 3 ) a strong increase in axon diameter . Hurd and Saxton reported that at sites of axonal swellings , individual axons increase in diameter up to 10-fold within a micron [13] . We performed ultrastructural analysis to address whether axons were swollen upon KhcN262S overexpression ( Figure 4A ) . An example of a typical cross-section of nerves of control larvae and those expressing KhcN262S is shown in the left panel of Figure 4A . The median axonal diameter was markedly increased in KhcN262S-expressing larvae ( w1118: 0 . 25 µm , n = 320; KhcN262S: 0 . 36 µm , n = 301 ) ( Figure 4B ) . Axons of control larvae contain microtubules ( Figure 4A , cyan arrowhead ) , cargos of fast axonal transport such as mitochondria ( Figure 4A , green arrowhead ) , and clear vesicles . As reported earlier for khc mutants [13] , axonal swellings observed in KhcN262S-overexpressing larvae additionally contained large dark-staining organelles , including multivesicular bodies ( Figure 4A , purple arrowhead ) , dark prelysosomal vacuoles ( PLVs ) ( Figure 4A , red arrowhead ) , and autophagosomes ( Figure 4A , dark blue arrowhead ) . To further validate the presence of lysosomal organelles in axonal swellings we expressed LAMP-GFP a marker for late endosomes and lysosomal compartments in motoneurons [25] . While no strong LAMP-GFP fluorescence was detected in control larvae , LAMP-GFP was strongly enriched in axonal swellings ( Figure 4C , arrowhead ) of KhcN262S-expressing larvae . The autosomal and autolysosomal marker ATG8-mrfp [26] also localized in axonal swellings ( Figure 4D , arrowhead ) . The accumulation of PLVs can be triggered by impairments in the retrograde transport of small prelysosomal organelles , which then fuse and mature , giving rise to the PLVs observed in swellings [13] . Alternatively , stress-driven autophagy of the cytoplasm [27] might further contribute to the formation of PLVs [13] . The fact that the swellings are positive for ATG8 is consistent with the hypothesis that stress-driven autophagy of the cytoplasm might contribute to the formation of PLVs observed in electron microscopy [13] . We sought to use in vivo analysis of axonal transport to estimate disturbances in the delivery of cargo to synapses . The frequency at which organelles are delivered to synapses can be predicted by measuring cargo flux , i . e . the number of organelles that pass a defined cross-section of a nerve in a given time interval . Transport velocity allows estimation of how long it takes an organelle to reach its destination . A slow velocity might indirectly cause crowding in the axon . Upon a 50% reduction of cargo velocity , a 100% increase of cargo density is necessary to obtain the same flux . Both genetic deletion of KIF5A [28] and ectopic expression of KIF5AN256S [29] reduced anterograde and retrograde flux of neurofilaments in cultured mouse cortical neurons . Although no effects of overexpression of KIF5AN256S on anterograde velocity were reported [29] , deletion of KIF5A ( KIF5A−/− ) reduced both maximum and average velocities of neurofilament transport [28] . Neurofilaments were not depleted from distal axons upon overexpression of KIF5AN256S [29]; neither axonal swellings nor increased apoptosis was reported [29] . In cultured KIF5A−/− motoneurons [30] , anterograde and retrograde transport velocities of mitochondria were reduced [30] compared with those of controls ( KIF5A+/+ ) . Effects on mitochondrial flux had not been investigated to date . We thus sought to determine the effects of deleting KIF5A on cargo flux . Anterograde ( KIF5A+/+: 0 . 10±0 . 022 min−1; KIF5A−/−: 0 . 06±0 . 015 min−1 , p = 0 . 045 ) but not retrograde flux of mitochondria ( KIF5A+/+: 0 . 08±0 . 018 min−1; KIF5A−/−: 0 . 08±0 . 017 min−1 , p = 0 . 93 ) was affected by loss of KIF5A . The number of stationary mitochondria detected within a 20 µm segment of the axon ( KIF5A+/+: 3 . 11±0 . 75; KIF5A−/−: 2 . 83±0 . 429 , p = 0 . 958 ) was not altered . Reductions in flux might be directly caused by reductions in transport velocity , or might be attributable to secondary defects . The 50% reduction of anterograde velocity that had been reported for KIF5A−/− motoneurons [30] fully explains the observed 44% reduction in anterograde flux that we observed . Although loss of khc [12] resulted in the reduction of retrograde flux rates in Drosophila , no impairment was observed in the KIF5A−/− motoneuron culture model . Our measurements were performed in motoneurons isolated at day E12 . 5 and assessed at day 4 in vitro , an early developmental time point that corresponds to an early stage of pathological progression at which no retrograde depletion of cargo would occur . We suggest that secondary defects , e . g . impaired microtubule stability , the formation of axonal traffic jams , or the distal depletion of mitochondria , might contribute to reductions in retrograde flux observed in khc deficient larvae [12] . We were thus interested in performing in vivo imaging to address the effects of the expression of Khcwt+N262S on anterograde and retrograde cargo transport in motoneurons of the Drosophila in vivo model at a time point at which behavioral impairments can be observed . Data obtained in Khcwt+N262S-expressing larvae are therefore of particular importance . These larvae allowed us to quantify , for the first time , the effects of expression of mutated Khc at physiologically relevant ratios on axonal transport in the context of an intact nerve . In Khcwt+N262S-expressing larvae , stochastically more than 25% of the Khc motors are expected to be Khcwt-homodimers . We did not observe a significant change in the velocity of mitochondria in either direction in any of the investigated genotypes . We did , however , detect a strong reduction in both anterograde and retrograde flux in both KhcN262S and Khcwt+N262S-expressing larvae . Khcwt-expressing larvae did not show any changes ( Figure 5A–5F; Videos S4 , S5 , S6 , S7 ) . Ebbing and colleagues [7] reported that in vitro velocities of purified KIF5A constructs were reduced more than two-fold upon mixing wild-type and N256S-mutant kinesin at a stoichiometric ratio of 1∶1 . The authors further assumed that kinesin cargo vesicles are moved by 5 to 8 motors [7] . Under these conditions , each organelle is expected to have a high probability of being attached to at least one mutant motor , leading to slower motility and shorter run lengths . The fact that we did not observe slower mitochondria suggests that the assumptions used to extrapolate single-molecule measurements to organelle transport in a cellular environment might be oversimplified . Alternatively , the experimental approach chosen to measure transport in vivo might be flawed . Bleaching is routinely used to quantify mitochondrial transport in Drosophila [12] , [18] . To exclude an influence of the bleaching procedure on our results , we sought to compare flux and velocities obtained before and after bleaching . We are not aware of a study that experimentally validates that transport velocities are not affected by the bleaching procedure . Theoretically , slow mitochondria might not enter the bleached region during the analyzed time interval . Thus , they might be excluded from the analysis , resulting in a biased analysis due to an inappropriate selection of fast-moving mitochondria . By comparing the mitochondrial flux in bleached and in non-bleached nerve segments , we could show that bleaching has an effect on flux rates; a higher flux is observed when analysis is performed after bleaching ( Figure 6A , 6B ) . This observation is best explained by the fact that bleaching allows for better visualization of moving mitochondria , which are less likely to be obscured after stationary mitochondria have been bleached . Both anterograde and retrograde flux is affected to the same degree by the method chosen . Thus , the ratio of retrograde to anterograde transport flux is not affected by the experimental procedure ( Figure 6C ) . We observed no effect of bleaching on transport velocities ( Figure 6D , 6E ) . To further confirm our results , we additionally performed a comparison of transport velocities obtained from two non-bleached control genotypes ( D42>w1118; D42>Khcwt ) and two non-bleached mutant genotypes ( D42>KhcN262S; khc−/− ) . No significant reduction of anterograde or retrograde transport velocities was detected in any of the investigated phenotypes ( Figure 6D , 6E ) . As no effect of bleaching on velocities could be observed , we suggest the use of bleaching when quantifying mitochondrial transport in Drosophila larvae . We were next interested in further investigating the cause of the reduced mitochondria flux . The reduced flux might be directly caused by impairments in axonal transport or by depletion of mitochondria in the cell body or near synapses . No obvious reduction in mitochondrial abundance in motoneuron cell bodies was detected ( Figure 7A ) . Thus , the observed reduction in anterograde flux is likely caused by impaired transport of mitochondria . Although the size of mitochondria was not affected by expression of mutated Khc , a strong reduction in the number and density of mitochondria at neuromuscular junctions ( NMJs ) 6/7 in segment 2 was detected in KhcN262S-expressing larvae ( Figure 7B–7E ) . A trend toward a reduced mitochondrial number ( p = 0 . 07 ) and density ( p = 0 . 08 ) in Khcwt+N262S-expressing larvae was observed ( Figure 7C–7E ) . Thus , reductions in the retrograde flux might be the result of impaired retrograde axonal transport or of reduced abundance of mitochondria at the synapse , or a combination of both effects . Next , we were interested in studying the structure and function of NMJs in more detail . Behavioral experiments in Khcwt+N262S-expressing larvae and data obtained in khc null mutants [13] predict strong defects at posterior segments , whereas anterior segments should be less affected . Quantification of both the NMJ area and the synapse number ( postsynaptic glutamate receptor fields ) revealed that this is indeed the case ( Figure 8A–8E ) . Overexpression of KhcN262S or Khcwt+N262S led to a strong reduction in the area of NMJs 6/7 in segment A5 but not in segment A2 ( Figure 8A–8C ) . Affected NMJs are furthermore characterized by inhomogeneity in anti-HRP staining ( Figure 8A , arrowheads ) . No significant reduction in the number of synapses was detected in anterior segment A2 ( Figure 8D ) . Overexpression of KhcN262S or Khcwt+N262S caused a significant reduction , however , in the number of synapses ( Figure 8E ) in posterior segment A5 . We next sought to address whether reduced axonal transport does limit the supply of NMJs with SVs and active zone proteins . To this aim , we used CSP and DV-Glut as markers for SVs and Brp as a marker for AZs . All three proteins are present in axonal swellings ( Figure 3B , 3D , 3H ) . We could confirm that the abundance of both AZ ( Figure 9A , 9B ) and SV ( Figure 9C–9F ) proteins is reduced at the NMJ . SV proteins are inhomogenously distributed in KhcN262S-expressing larvae . While few boutons stain intensively for CSP and DV-Glut ( Figure 9C , 9E arrowheads ) , other boutons display a weaker staining intensity ( Figure 9C , 9E arrows ) . There is a strong correlation between the inhomogeneity observed in the staining for HRP and SV proteins . This inhomogenous distribution might resemble defects in the delivery of SV , in endo-/exocytosis , or in membrane trafficking . Thus , we were interested in addressing functional impairments of the NMJ in detail . To this aim , we recorded postsynaptic potentials by using current clamp recordings at muscle 6 in segment A4 . The evoked excitatory junction potentials ( eEJPs ) of KhcN262S-expressing larvae were drastically decreased in amplitude and had an increase in the half-width time ( Figure 10A , 10B , 10C ) . The increased half-widths in KhcN262S- and Khcwt+N262S-expressing larvae might be caused by impairments in the synchronization of vesicle fusion with the arrival and spread of the action potential to all release sites [23] . Loss of Brp - an active zone protein that has been shown to be important for establishing close proximity between SV and release sites - leads , in like manner , to an increase in the half-width of evoked excitatory junctional currents [23] . In contrast to eEJPs , the amplitude of miniature excitatory junction potentials ( mEJPs ) in response to single , spontaneous vesicle fusion events was slightly - but not significantly – increased in both KhcN262S- and Khcwt+N262S-expressing larvae ( Figure 10D , 10E ) . This might represent a postsynaptic compensation presynaptic defect . Indeed , although there was no significant difference in eEJPs between Khcwt+N262S-expressing larvae and controls , the former group revealed a significant reduction in quantal content ( Figure 10F ) . As quantal content is a measure of the number of vesicles released per presynaptic action potential , it is better suited for characterizing presynaptic defects than eEJP size . As HSP is a neurodegenerative disorder characterized by distal axonopathy , we were interested in whether we could observe any signs of synapse degeneration in our models . Both KhcN262S- and Khcwt+N262S-expressing larvae showed pathological alterations in neuronal membrane organization ( Figure 8A; Figure 9A , 9C , 9E ) , as well as reduced abundance and altered distribution of SV proteins ( Figure 9C , 9E; Figure 11A–11D ) . However , typically the complete absence of the SV protein synapsin from parts of the NMJ was not detected in KhcN262S-expressing larvae ( Figure 9C , 9E; Figure 11C , 11D ) . These data are consistent with the reduction in abundance and inhomogeneous distribution of CSP and DV-Glut ( Figure 9C , 9E ) observed in KhcN262S-expressing larvae . Synaptic footprints [31] , areas of the NMJ that are , after a retraction of the nerve , positive for postsynaptic marker proteins ( Dlg or GluRIIC ) , but negative for presynaptic marker proteins , are commonly scored by using either synapsin [31]–[33] or Brp [34] as a presynaptic marker . The absence of a single presynaptic marker protein is a clear indication of pathological alterations at the NMJ . It is not sufficient , however , to prove that a nerve ending has retracted . We thus defined only the simultaneous absence of an SV marker ( synapsin ) and of the presynaptic membrane ( HRP ) as a retraction event . Using this more conservative assay , retraction was seldom detected . Dystrophic boutons characterized by a strong reduction in the intensity of SVs , in combination with an inhomogeneous HRP signal , were , however , frequently detected in animals expressing mutant Khc . To quantify the degree of neurodegenerative pathological alterations at the NMJs , we used a scoring system that assessed the frequency of retractions , the occurrence of dystrophic boutons , and the presence of minor pathological alterations such as weaker staining for SV ( compare Figure S1A–S1F and Text S1 ) . Using this scoring system , we detected a significant degree of neurodegenerative alterations in larvae expressing mutant Khc ( Figure 11E ) . A strong HRP staining at a subset of boutons at the NMJs of larvae expressing mutant Khc suggested the local accumulation of membrane rich organelles . As autolysosomal organelles were detected in axonal swellings , we sought to address whether the observed strong HRP signal might be indicative of increased autophagy at the NMJ . Indeed , ultrastructural analysis of larvae overexpressing Khcwt+N262S indicated that PLVs ( red arrowhead ) , autophagosomes ( dark blue arrowhead ) , and multivesicular bodies ( purple arrowhead ) are frequently present in mutant but not in control NMJs ( Figure 12A and 12BI–12BIII ) . Next , we questioned whether the observed degeneration represents a classic distal synaptopathy or whether it is preceded by the loss of the motoneuron cell body . To this aim , we analyzed motoneuron cell bodies in 4-day-old early-mid L3 larvae , the stage at which degeneration was observed at the NMJ . No substantial motoneuron loss was detected in the subset of motoneurons positive for eve in KhcN262S-expressing larvae when compared with the control group ( Figure 12C ) . To monitor cell death in all motoneurons we used an antibody that allows visualizing the activation of the putative initiator caspase DRONC [35] . Using this antibody , which is commonly used to quantify dying cells in Drosophila ( for review see [35] ) , we found no evidence for increased apoptosis of motoneurons in mid L3 KhcN262S-expressing larvae ( Figure 12D ) . The fact that we did not observe substantial motoneuron cell loss in mid-L3 KhcN262S-expressing larvae , a stage when degeneration of synapses was already pronounced , is consistent with the concept that HSP is primarily caused by synaptopathy and axonopathy , whereas motoneuron cell loss is neither causative for HSP nor an early feature of the pathological process .
Human genetics suggest that autosomal-dominant SPG10 is not caused by haploinsuffiency . The above presented Drosophila model further supports this hypothesis . While no difference in the locomotion speed between wild-type larvae and larvae lacking one copy of khc were observed , larvae ectopically expressing KhcN262S were severely impaired . The observed behavioral impairments are both qualitatively and quantitatively similar to impairments observed in khc null larvae . These results are consistent with KhcN262S acting as an antimorph or a neomorph . Antimorphic mutations act in opposition to the normal gene function . Thus , they are also referred to as dominant negative mutations [36] . The phenotypic severity of an antimorphic mutation will be decreased by increasing wild-type gene dosage [36] . A neomorphic mutation leads to a change in the nature of the gene resulting in a dominant gain of function . This function , which is not produced to a relevant degree by the native gene , is often toxic . Increasing wild-type gene dosage will not reduce the phenotypic severity of a neomorphic mutation , as the newly gained function is by definition different from the normal gene function [36] . Animals ectopically expressing KhcN262S alone were more severely affected than animals expressing KhcN262S in combination with an additional copy of Khcwt . We thus conclude that KhcN262S is an antimorphic mutation: KhcN262S causes loss of kinesin-1 function via a dominant-negative mechanism . We propose that all observed pathological changes are downstream of a loss of kinesin-1 function and cannot be attributed to more generalized toxicity of mutant Khc or kinesin-1 complexes containing mutant Khc . In human patients , 25% of the Khc motors are expected to be Khcwt-homodimers . Our results suggest that these should be capable of transporting synaptic cargo . Two models have been proposed to explain how molecular motors coordinate cargo transport ( for review , see [37] ) . The tug-of-war model assumes that the direction of movement is the result of the dynamic competition of opposing motors . The simultaneous action of two motors thus exerts a stretching force on the particle . Loss of motors responsible for transport in one direction should increase transport velocities in the opposite direction . While the direct opposing action of dynein and the Dictyostelium kinsesin-3 family member Unc-104 could be experimentally validated [38] , most frequently , the loss of motors responsible for transport in one direction leads to transport disturbances in both directions [30] , [39] . This is best explained by a model in which the activity of opposing motors is controlled by coordination complexes that are regulated such that only one set of motors is active at any given time . These two models are , however , not mutually exclusive . Both models offer distinct advantages for molecular motors . Ideally , coordinated action of opposing motors might allow for faster , more energy-efficient transport , but simultaneous binding of opposing motors has been proposed to decrease the probability of detachment from microtubules , resulting in increased processivity of movement [40]–[42] . Thus , it is most probable that in most cellular environments cargo is transported by the coordinated , simultaneous action of opposing motors . The exact balance between coordinated inactivation of opposing motors and their active role as a stabilizing “dragging force” might vary for different cargos , developmental time points , stages of disease progression , and distinct cell types . Thus , it is of no surprise that studies investigating mutation in molecular motors performed in different model systems lead to seemingly opposing results . In Drosophila , mitochondrial transport velocities are affected neither by the deletion of khc [12] nor by overexpression of KhcN256S . These observations are in accordance with data obtained by measuring neurofilament transport in cultured mouse cortical neurons ectopically expressing KIF5AN256S [29] . Yet how can these results be reconciled with the observation that maximum and average velocities of mitochondrial [29] and neurofilament [28] transport were reduced in neurons isolated from KIF5A−/− mouse embryos ? The loss of KIF5A might be partially compensated by KIF5B or KIF5C . KIF5B and KIF5C might be less effective , however , in transporting mitochondria , thus reducing both average and maximum transport velocity . While no compensatory up-regulation of KIF5B or KIF5C was detected in KIF5A−/− mice [30] , indirect evidence nonetheless suggests that residual mitochondrial transport observed in KIF5A−/− motoneurons might be driven by KIF5B or KIF5C . First , both mitochondria and neurofilaments are actively transported in the anterograde direction [28] , [30] . Thus , they must be bound to an anterograde motor . Second , KIF5C and KIF5B have been shown to be important for mitochondrial transport [43] . In KIF5B−/− cells , mitochondria are clustered around the nucleus rather than being appropriately dispersed throughout the cell . This defect can be rescued by ectopic expression of KIF5A , KIF5B , or KIF5C , highlighting the fact that any of these motors are capable of transporting mitochondria [43] . While the frequency of neurofilament transport is reduced by 75% in KIF5A−/− neurons , KIF5C or KIF5B overexpression is sufficient to partially rescue this defect [28] . We thus propose that cellular quality control mechanisms ensure that cargos are equipped with a minimal number of molecular motors . Thus , obvious defects such as the genetic deletion of KIF5A are detected and partially compensated by targeting similar motors to cargos awaiting initiation of transport . Transport driven by these alternative motors might be slower and less efficient [28] , [30] . Still , a basal cargo flux can be obtained despite the complete absence of one molecular motor type [28] , [30] . Mutations in KIF5A that impact neither the stability of the protein nor its ability to interact with regulatory complexes might not be detected by this cellular quality control system . Thus , cargos might be loaded with defective motors that cause problems after initiation of transport . These secondary problems might include a drop of transport flux [28] , [29] . The phenotypic strength of defects might therefore depend on the expression level as well as the nature of the mutation . Thus , observed defects might be either less severe ( KIF5AN256S [29] ) or more severe ( head-less KIF5A , [28] ) than loss of KIF5A [28] . KhcN262S-expressing larvae exhibit the characteristics of classical distal degeneration . No human SPG10 autopsy reports have been published till date , but studies of SPG4 cases [1] suggest a “dying back” axonopathy as probable disease mechanism in HSP . Synapse and axon loss may therefore be a primary step in the pathophysiological manifestation of SPG10 , while neuronal loss may occur only later . Could protection of the cell bodies be used as a suitable treatment strategy for HSP ? While little data is available on the treatment of HSP patients , treatment strategies for Amyotrophic lateral sclerosis ( ALS ) , a clinically related disease , have been explored in more detail . The neurodegenerative disease ALS is characterized by the progressive loss of motoneuron in the brain and the spinal cord [44] . Pathological changes occur in ALS patients and animal models - similar to our observations in the Drosophila HSP model - first at the NMJ and are followed by axon and neuronal loss [45] , [46] . Treatment strategies aiming at prevention of motoneuron loss lead only to limited success in ALS models and human patients [44] . Thus , treatment that aims at delaying motor impairments during the progression of ALS is currently favored [44] . We suggest that preservation of synapses and axons will be an important requirement for a successful therapeutic intervention of both HSP and ALS . Impaired neurofilament transport has been implicated in the pathogenesis of SPG10 [29] . The human genome contains 3 genes encoding the heavy chains of conventional kinesin: the neuronally expressed genes KIF5A and KIF5C and the ubiquitously expressed gene KIF5B . Neurofilaments are transported by KIF5A and KIF5C [28] , [29] . Impairments in neurofilament transport due to impaired KIF5A function might thus explain why only mutations in KIF5A , but not KIF5B , have been identified as a genetic cause for HSP . The Drosophila genome , on the other hand , does not contain a neurofilament gene [47] . Electron microscopic studies concluded that neurofilaments are absent in all arthropods [48] and yet ectopic expression of mutant Khc leads to the formation of axonal swellings and HSP-like pathological changes in Drosophila . Conversely , the hypothesis that impairments in neurofilament transport represent an important cellular cause of HSP has been further supported by reports that swellings in SPG4 patients and the SPG4 mouse model were positive for neurofilaments [49] . These swellings also contained , however , multiple other cargos , including mitochondria , the amyloid precursor protein , tubulin , and tau [49] . In our study , both kinesin-1 and kinesin-3 cargos [24] , as well as lysosomal and autophagic organelles , accumulated in swellings . This suggests that eventually all kinds of cargo might become trapped in axonal swellings . Future studies involving the use of in vivo imaging in mouse and Drosophila models will be needed to shed light on the formation of swellings . It is therefore important to identify cargos that accumulate first within a swelling . The cargo , which causes the formation of the swellings , should accumulate prior to other cargos . A more detailed understanding of the temporal sequence of cargo accumulation and an increase in the understanding of impaired neurofilament transport will be instrumental to further decipher common pathological hallmarks of HSP . It is to be hoped that this will aid in the design of successful treatment strategies .
Site-directed mutagenesis was used to introduce the amino acid exchange N262S in a full-length wild-type Drosophila khc cDNA ( SD02406 ) which was next inserted into a modified pUAST attB vector . Details are described in Text S2 . Flies were maintained at 25°C on standard fly medium seeded with yeast . For experiments , flies were raised at 18°C , 25°C , or 29°C . For fly strains used in this study , see Table S1 . Transgenic stocks UAS-khcwt and UAS-khcN262S were created by BestGene using integrase mediated site-specific transgenesis at cytological position 86F ( Fly strain BDSC 23648 ) . Proteins were extracted from whole fly heads using 65 mM Tris ( pH 6 . 8 ) , 5% ( w/v ) SDS , 1× Protease Inhibitor ( Roche ) buffer . Samples were separated on 7 . 5% SDS gels and transferred to nitrocellulose membranes . For antibodies used in this study , see Tables S2 and S3 . Third instar larvae were dissected in chilled Ca2+-free HL3 solution and fixed in 4% formaldehyde in PBS . For antibodies used in this study , see Tables S2 and S3 . Larval preparations were mounted in Vectashield ( Vector ) . Images were captured using a Zeiss LSM 710 confocal microscope with the following settings unless otherwise noted: Objective: 40× plan Apochromat , 1 . 3 N . A . ; Voxel Size: 100 nm×100 nm×500 nm; pinhole: 1 AU , average: 2–4 . Images used for illustration purposes were processed as follows: ( 1 ) A Gaussian filter ( radius = 2 ) was applied to the raw data stack . Brightness and contrast were appropriately adjusted . The relevant slices of the modified stacks were maximum-projected . Projected images were scaled by 2 , and gamma adjustment ( gamma = 0 . 75 ) was applied . ImageJ 1 . 41o , 1 . 44p , 1 . 46r or 1 . 45s ( US National Institutes of Health; http://rsb . info . nih . gov/ij/download . html ) was used to process and analyze images . For quantification of Glutamate receptor fields , Delta 2D software ( Decodon GmbH , Germany ) was used . In vivo imaging was essentially performed as previously described [50]–[52] , using a Zeiss LSM 710 confocal microscope equipped with a 40× Plan Apochromat Objective ( 1 . 3 N . A . ) . For better visualization of moving mitochondria , all mitochondria in a 20 µm segment of the nerve were bleached . This allowed for easy visualization of moving particles passing through the bleached region . Next , confocal Z-stacks ( Z-planes: 10; voxel size: 100 nm×100 nm×1500 nm , pinhole: 1 . 6 AU , average: 2 ) were recorded at maximal speed corresponding to 100 stacks per 426 seconds . For details on the generation of kymographs , see Text S1 . Third instar larvae were dissected , rinsed , and transferred into the recording chamber . A fixed-stage upright microscope ( Model BX51WI with 40× water immersion lens; Olympus Optical , Tokyo , Japan ) was used to visualize the nerve and the muscles . Intracellular current clamp recordings were performed in HL3 solution with 1 mM extracellular Ca2+ at 19°C . Evoked excitatory junction potentials ( eEJPs ) and spontaneous miniature excitatory junction potentials ( mEJPs ) were obtained from muscle 6 segment A4 with an Axoclamp 900A amplifier ( Axon Instruments , Union City , CA ) , digitized ( Digidata 1440A , Axon Instruments , Union City , CA ) , recorded at 10 kHz ( pClamp 10 , Axon Instruments , Union City , CA ) , and analyzed using AxoGraph X software . Sharp , bee-stinger-shaped glass microelectrodes filled with 3 M KCl and a resistance between 10 and 20 MΩ were used . Cells with resting potentials between −60 and −70 mV and input resistance >4 MΩ were selected for analysis . For stimulation , the cut end of the segmental nerve was pulled into a fire-polished suction electrode ( 6–8 µm inner diameter ) , and brief ( 300 µs ) depolarizing pulses were passed at 0 . 1 Hz ( ISO-STIM 01D , NPI Electronics , Tamm , Germany , stimulus generator and isolation unit ) . The amplitude of the pulse was set to about 7V , which results in the stable recruitment of both innervating motoneurons . It corresponds to 1 . 5 times the amplitude needed to recruit both motoneurons innervating muscle 6 . For each eEJP and mEJP average , 15 eEJPs and 120 s of mEJP recordings were used for subsequent analysis . Larval fillets were fixed with 4% PFA ( in PBS ) for 10 min at room temperature followed by fixation in 2 . 5% glutaraldehyde ( in PBS ) overnight at 4°C . Postfixation was done with 1% osmium tetroxide in 100 mM phosphate buffer , pH 7 . 2 , for 1 h on ice . Larval fillets were rinsed with water , treated with 1% aqueous uranyl acetate ( UA ) for 1 h at 4°C , dehydrated through a graded series of ethanol concentrations , and stored in liquid Epon overnight . Next , larval fillets were pinned on a dissection pad . Muscles 4 of segment 4 were dissected with sharp insect pins , embedded in Epon , and polymerized for 48 h at 60°C . Ultrathin sections were stained with UA and lead citrate and viewed in a Philips CM10 electron microscope . Living L2 larvae were transferred to aluminum platelets with a 150 µm recess containing 1-hexadecene as an external nonpenetrating filler . The platelets were sandwiched with platelets that had no recess and cryofixed with a high-pressure freezer ( Bal-Tec HPM 010 , Balzers , Liechtenstein ) . Larvae were freed from external hexadecene under liquid nitrogen and then transferred to 2 ml microtubes with screw caps ( Sarstedt , #72 . 694 , Germany ) . As a freeze substitution medium , we used a 2% osmium tetroxide solution in anhydrous acetone , supplemented with 25 µl of 20% methanolic UA solution to give a final UA concentration of 0 . 5% . The freeze substitution was carried out in a Leica AFS-2 with the samples kept at −90°C for 27 h , −60°C for 6 h , and −40°C for 6 h . The temperature increase between steps was set to 10°C/h . At −40 h , glutaraldehyde from a 25% solution in water ( EMS #16530 , Electron Microscopy Sciences , Fort Washington , PA ) was added to give a final concentration of 0 . 6% glutaraldehyde and 2% water . After 6 h at −40°C , the microtubes were placed on ice for another 1 h . Samples were then washed 3 times with acetone and infiltrated with Epon at room temperature in a series of increasing Epon concentrations in acetone ( 30% , 60% , 90% , 2× 100% Epon each for 1 h , with the second 100% Epon change continuing overnight on a rotating wheel ) . After embedding , the Epon samples were polymerized for 48 h at 60°C . Ultrathin sections were stained with UA and lead citrate and viewed in a Philips CM10 electron microscope . Micrographs were taken on EM-film ( Maco ES 208 , Hans O . Mahn GmbH & Co KG , Stapelfeld , Germany ) . Alternatively , sections were imaged by using a FEI-Tecnai Spirt , 120 kV electron microscope equipped with a Gatan USC 4000 camera . Thoraxes were dissected by careful removal of heads , wings , limbs , and abdomen parts , prefixed at 4°C overnight ( 4% PFA , 3% glutaraldehyde , 0 . 1% sodium cacodylate ) , and postfixed with 1% osmium tetroxide for 3 h at 4°C . Next , thoraxes were washed with 30% , 50% , 70% , and 100% ethanol for 10 min each followed by washing with 100% acetone and 3∶1 acetone∶Epon for 1 h . Thoraxes were then immersed in 1∶1 acetone∶Epon and 100% Epon for 24 h . The Epon was polymerized for 48 h at 60°C . Semi-thin sections ( 2 µm ) were prepared on a Reichert-Jung Supercut 2050 microtome with glass knifes . The semi-thin sections were stained with Toluidine blue solution ( 0 . 5% Toluidine blue O [C . I 52040 , Roth] in 1% [w/v] disodium tetraborate buffer ) for 1 minute and then washed under running water . The semi-thin sections were documented on a Zeiss Imager . Z1m microscope by using a 20× Zeiss Neofluar , 0 . 5 N . A objective . All washing , fixation , or staining procedures were performed at room temperature unless otherwise noted . Flies were raised at 18°C . Offspring were collected on the day of eclosion and 15–20 male flies transferred to vials containing standard fly media . The flies were transferred to fresh fly media every 3 days . A Kaplan-Meier plot was used for depicting survival curves . Flies were raised at 18°C . Emerged male flies were split into two batches of 50 flies within 24 h after eclosion . One batch of flies was raised at 18°C and the other at 29°C to induce the expression of either Khcwt or KhcN262S for 16 days . Motor function of 16-day-old flies was monitored by analyzing their ability to climb 6 cm at the wall of a vertical plastic tube within 15 s . Fifty flies from each genotype were analyzed . A successful trial was scored with 1 and a nonsuccessful trial with 0 . Each fly was allowed to climb three times and the average climbing score per fly was calculated . To monitor locomotion behavior , we placed individual larvae on a thin slice of apple juice agar . Locomotion was examined at 25°C at 70% humidity by using a DCM510 ( ScopeTek , P . R . China ) camera integrated in a custom-built stereomicroscope . Larval locomotion was recorded at a frame rate of 30 fps for 5 min . The videos were then converted into avi format by using a Prism Video Converter , v 1 . 61 ( NCH Software Inc . , Australia ) . Next , images were cropped and compressed by using VirtualDub 1 . 9 . 10 ( http://www . virtualdub . org/ ) . To measure locomotion speed , we placed up to 200 larvae on a 15×15 cm agar plate and filmed for 10 min . Locomotion and size of the larvae were analyzed with the custom-built software Animaltracer . This software was developed by us on the basis of the MATHLAB software package Worm Tracker & Track Analyzer ( Department of Molecular and Cellular Physiology at Stanford University ) [53] . This algorithm can be divided into two parts , the larval tracker and the track analyzer . The larval tracker identifies and tracks individual larvae within a movie . The track analyzer analyzes the movies and returns the size and the velocity of single larvae . For comparative analysis of the different genotypes larvae within a certain size range were grouped . Average locomotion speed is calculated for this size group for every movie . Larvae that touched each other were automatically excluded from analysis . Larvae whose velocity was less than 10% of average velocity of the respective genotype and size group were likewise excluded from analysis . A minimum of six movies per genotype were analyzed . For all further statistical analysis , n was defined as the number of movies analyzed . KIF5A+/− mice were obtained from MMRRC ( Mutant Mouse Regional Resource Centers , University of California , Davis , USA ) . Mouse embryonic motoneuron culture , staining of mitochondria , and time-lapse imaging were performed as previously described [30] . Flux of mitochondria was measured in a 20 µm segment of motoneuron axons . The number of mitochondria passing two defined cross-sections , both in the anterograde and the retrograde direction , were counted in a time interval of 30 minutes . The flux is the average of these two measurements . Mitochondria that moved less than 5 µm within the 20 µm segment were classified as stationary . 18 KIF5A+/+ and 23 KIF5A−/− axons from four independent experiments were analyzed . All animal work in this study were approved by the German Government ( Regierungspräsidium Tübingen ) and the University of Tübingen . Statistical tests were performed with the software PAST . exe ( http://folk . uio . no/ohammer/past/index . html ) unless otherwise noted . Sample errors are given as standard deviation ( s . d . ) and standard error of the mean ( s . e . m ) . Data were first tested for normality by using the Shapiro-Wilk test ( α = 0 . 05 ) . Normally distributed data were analyzed either by student's t-test ( two groups ) or by a one-way analysis of variance followed by a Tukey-Kramer post-test for comparing multiple groups . Non-normal distributed data were analyzed by using either a Mann-Whitney test ( two groups ) or a Kruskal-Wallis H-test followed by a Dunn's test for comparisons between multiple groups . Differences in survival were determined by the Mantel-Cox test using Prism . The p values obtained from the Mantel-Cox test were corrected for the total number of comparisons made . Statistical tests for analyzing axonal transport in KIF5A-deficient motor neurons were performed with IBM SPSS Statistics , Version 20 . The following alpha levels were used for all tests: * p<0 . 05; ** p<0 . 01; *** p<0 . 001 . | Hereditary spastic paraplegias ( HSPs ) comprise a group of inherited neurological diseases . The main feature of HSP is progressive stiffness of the lower limbs due to a dysfunction of nerve cells . We study HSP type 10 , which is caused by mutations in the neuronal motor protein KIF5A . HSP type 10 is inherited in an autosomal-dominant manner , which means that patients have a normal and a mutated copy of the KIF5A gene . KIF5A plays an important role in neuronal function: it transports cargos to the synapse that can be up to 1 m from the cell body . We use the fruit fly as a model to investigate the role of mutations in KIF5A . Our fly model replicates a central feature of HSP: muscles that are activated by nerve cells that have long cellular processes are more severely impaired . We now address why one mutated copy of KIF5A is sufficient to cause HSP . To date , it has been thought that patients might have HSP because they have insufficient functional KIF5A or because mutated KIF5A disturbs the function of normal KIF5A . We provide evidence for the latter possibility . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"model",
"organisms",
"neurological",
"disorders",
"neurology",
"biology",
"molecular",
"cell",
"biology",
"neuroscience",
"genetics",
"and",
"genomics",
"clinical",
"genetics"
] | 2012 | Spastic Paraplegia Mutation N256S in the Neuronal Microtubule Motor KIF5A Disrupts Axonal Transport in a Drosophila HSP Model |
The Hippo signaling pathway has a conserved role in growth control and is of fundamental importance during both normal development and oncogenesis . Despite rapid progress in recent years , key steps in the pathway remain poorly understood , in part due to the incomplete identification of components . Through a genetic screen , we identified the Drosophila Zyxin family gene , Zyx102 ( Zyx ) , as a component of the Hippo pathway . Zyx positively regulates the Hippo pathway transcriptional co-activator Yorkie , as its loss reduces Yorkie activity and organ growth . Through epistasis tests , we position the requirement for Zyx within the Fat branch of Hippo signaling , downstream of Fat and Dco , and upstream of the Yorkie kinase Warts , and we find that Zyx is required for the influence of Fat on Warts protein levels . Zyx localizes to the sub-apical membrane , with distinctive peaks of accumulation at intercellular vertices . This partially overlaps the membrane localization of the myosin Dachs , which has similar effects on Fat-Hippo signaling . Co-immunoprecipitation experiments show that Zyx can bind to Dachs and that Dachs stimulates binding of Zyx to Warts . We also extend characterization of the Ajuba LIM protein Jub and determine that although Jub and Zyx share C-terminal LIM domains , they regulate Hippo signaling in distinct ways . Our results identify a role for Zyx in the Hippo pathway and suggest a mechanism for the role of Dachs: because Fat regulates the localization of Dachs to the membrane , where it can overlap with Zyx , we propose that the regulated localization of Dachs influences downstream signaling by modulating Zyx-Warts binding . Mammalian Zyxin proteins have been implicated in linking effects of mechanical strain to cell behavior . Our identification of Zyx as a regulator of Hippo signaling thus also raises the possibility that mechanical strain could be linked to the regulation of gene expression and growth through Hippo signaling .
The Hippo pathway has emerged as an important regulator of growth during metazoan development , and its dysregulation is implicated in diverse cancers [1]–[3] . Hippo signaling is effected by transcriptional co-activator proteins , Yorkie ( Yki ) in Drosophila and YAP and TAZ in mammals [4] . Three interconnected , upstream branches of Hippo signaling have been characterized in Drosophila: Fat-dependent , Expanded-dependent , and Merlin-dependent [1]–[3] . These upstream branches converge on the kinase Warts ( Wts ) , which can phosphorylate Yki . Phosphorylated Yki is retained in the cytoplasm , whereas unphosphorylated Yki can enter the nucleus and , in conjunction with DNA-binding partners , promote the transcription of downstream genes . Upstream branches of Hippo signaling regulate both the activity of Wts and its abundance . Our understanding of many steps in Hippo signaling remains fragmentary , in part due to incomplete identification of pathway components . Here , we describe the identification of Zyx102 ( Zyx , FBgn0011642 ) as a novel component of Hippo signaling and characterize its role in the pathway . Fat is large cadherin that acts as a transmembrane receptor for one branch of Hippo signaling [1]–[3] , [5] . Fat-Hippo signaling influences the levels of Wts protein [6] . The molecular mechanism by which this is achieved is not understood , but dachs is genetically required for the influence of Fat on Wts levels , downstream gene expression , and organ growth [6]–[8] . Fat regulates the localization of Dachs to the sub-apical membrane: when fat is mutant , Dachs accumulates on the membrane around the entire circumference of the cell , and when Fat is over-expressed , Dachs is mostly cytoplasmic [7] . In imaginal discs and optic neuroepithelia , Dachs membrane localization is polarized within the plane of the tissue; this polarization reflects the graded expression of the Fat ligand Dachsous and the Fat pathway modulator Four-jointed [7] , [9] , [10] . The correlation of Dachs localization with Fat activity implicates Dachs regulation as a key step in Fat signaling , but how Dachs localization influences downstream events is unknown . Zyx is a Drosophila homologue of the vertebrate Zyxin , Lipoma preferred partner ( LPP ) , and Thyroid-receptor interacting protein 6 ( TRIP6 ) proteins [11] , [12] . These proteins have three conserved LIM domains at their C-terminus , and they have been implicated in both cytoskeletal and transcriptional regulation [13]–[15] . Gene-targeted mutations in murine Zyxin or Lpp have no significant effect on mouse development , presumably due to redundancy among family members [16] , [17] . Translocations involving LPP identified it as an oncogene involved in lipomas and other cancers [13] . In cultured cell assays , Zyxin and its paralogues can affect cell motility and actin polymerization and can localize to focal adhesions and adherens junctions [13] , [15] , [18] . Notably , Zyxin has been implicated as playing a key role in mechanotransduction , as its localization to focal adhesions can be influenced by the application of mechanical tension to cells in culture [18] . We report here that Zyx is an essential component of the Fat-Hippo signaling pathway , required for normal Yki activity and growth in Drosophila . Using genetic epistasis tests , we position the requirement for Zyx in between fat and wts . Binding studies show that Zyx protein binds to Dachs and binds to Wts in a Dachs-regulated manner . Our observations suggest a model in which the regulated localization of Dachs to the membrane regulates Zyx-Wts binding , which then promotes Wts degradation . Dachs is a myosin protein , and its myosin motor domain contributes to interactions with Zyx and Wts , which raises the possibility that additional myosins might regulate Zyx-Wts interactions in other contexts .
Reduction of Zyx in the developing wing disc , under nub-Gal4 control ( Figure S1A ) , results in adult flies with small wings ( Figure 1A–C , S ) . Similar phenotypes were observed using two different RNAi lines , although NIG-32018R3 ( RNAi-Zyx32018 ) , the line identified in our original screen , has slightly stronger phenotypes . Hippo signaling also regulates leg growth , and depletion of Zyx in developing legs results in shorter legs with fewer tarsal segments ( Figure S1I , J ) . In addition to observing similar phenotypes with two independent RNAi lines , confirmation that the phenotypes observed result specifically from reduction of Zyx was provided by the observation that over-expression of Zyx from a UAS transgene rescued the RNAi phenotypes ( Figure 1D , S ) . We also confirmed by Western blotting that that Zyx RNAi reduced Zyx protein levels ( Figure S1K ) . Many different genes and pathways affect organ growth . To investigate the potential connection between Zyx and the Hippo pathway , we examined the expression of downstream target genes in wing discs in which Zyx was depleted by RNAi . As downstream targets we employed reporters of expanded ( ex ) expression ( ex-lacZ ) and th expression ( th-lacZ , Diap1 ) . When Zyx was depleted from posterior cells using en-Gal4 , ex-lacZ , th-lacZ , and Diap1were all reduced ( Figures 2A , B , S2A ) . Hippo signaling regulates transcription by controlling the sub-cellular localization of Yki: activation of Hippo signaling promotes cytoplasmic localization of Yki , whereas inactivation of Hippo signaling allows nuclear localization of Yki , which corresponds to Yki activation [22] , [23] . Zyx RNAi reduced nuclear Yki . This effect was subtle at late third instar , when levels of Yki in the nucleus are already low , but was evident in younger wing discs , which have higher levels of nuclear Yki ( Figure 2C , D ) . The decreased expression of Hippo pathway target genes , together with the reduction in nuclear Yki , identifies Zyx as a regulator or component of the Hippo pathway . The Hippo pathway is generally thought of as a negative regulator of growth and gene expression , because most genes in the pathway act as tumor suppressors and negatively regulate the activity of Yki . Zyx , by contrast , is positively required for Yki activity and organ growth . To position the genetic requirement for Zyx within the Hippo pathway , we performed a series of epistasis tests . RNAi lines targeted against several different tumor suppressor genes within the pathway ( fat , ds , ex , wts , hpo , and mats ) , each of which phenocopy their respective mutants , were examined in combination with Zyx RNAi lines . The immediate upstream regulator of Yki is wts . Expression of a wts RNAi line under nub-Gal4 or en-Gal4 control is lethal at late third instar , but imaginal discs can be recovered and analyzed before lethality . Consistent with the expected de-repression of Yki , expression of wts RNAi resulted in upregulation of ex and Diap1 expression ( Figure 3A ) . This upregulation of ex and Diap1 was not suppressed by Zyx RNAi ( Figure 3B ) ; hence , wts is epistatic to Zyx . Wts activity is directly regulated by a kinase , Hippo ( Hpo ) , and a co-factor , Mats , and hpo and mats were also epistatic to Zyx ( Figure S3A–D ) . These observations imply that Zyx acts upstream of Wts . Upstream branches of Hippo signaling have been characterized in Drosophila as Fat-dependent , Ex-dependent , or Mer-dependent . In the developing wing , fat and ex make substantial contributions to Yki regulation , whereas Mer has a lesser role [6] , [24]–[27] . Thus , we investigated the relationship between the requirement for Zyx and those for fat and ex . Expression of fat or ex RNAi throughout the wing , under nub-Gal4 control , results in overgrown wings ( Figure 1E , I , S ) . Strikingly , the wing overgrowth phenotype associated with depletion of fat was suppressed by Zyx RNAi , resulting in adult wings of similar size to those of animals that only expressed Zyx RNAi ( Figure 1B , F , S ) . This epistasis of Zyx to fat was also visible at the level of target gene expression ( Figure 3D , E ) and the subcellular localization of Yki ( Figure 4G , H ) . Zyx is also epistatic to the Fat ligand ds ( Figures 1S , S1C , D ) . These observations imply that Zyx acts downstream of fat . The ex RNAi phenotype , by contrast , was only slightly affected by Zyx RNAi , as the wings of Zyx ex double RNAi animals remained overgrown ( Figure 1J , S ) . Moreover , ex was epistatic to Zyx for effects on downstream target gene expression and Yki localization ( Figure 4A–D , J , K ) . Together , these observations indicate that Zyx specifically affects Fat-Hippo signaling and has little effect on Ex-Hippo signaling . To refine our placement of Zyx within Fat-Hippo signaling , we examined requirements for Zyx relative to additional pathway components . dco encodes a kinase that phosphorylates the Fat cytoplasmic domain and participates in Fat-Hippo signaling [6] , [28] , [29] . The requirement for Dco within Fat signaling is uncovered by expression of an antimorphic isoform , Dco3 . Expression of Dco3 induces wing overgrowth ( Figure 1L ) [29] . This overgrowth is suppressed by Zyx RNAi , suggesting that Zyx acts downstream of dco ( Figure 1P , S ) . Like Zyx , dachs is required for normal wing and leg growth and acts genetically downstream of fat and dco but upstream of warts [6]–[8] . To examine the genetic relationship between Zyx and dachs , we took advantage of the observation that over-expression of Dachs can promote wing overgrowth ( Figure 1Q ) [7] . This overgrowth was completely suppressed by Zyx RNAi ( Figures 1S , S1G ) , as was the influence of Dachs over-expression on ex-lacZ expression ( Figure S4A , B ) . Thus , Zyx is required for Dachs-promoted activation of Yki . Over-expression of Zyx resulted in a mild wing overgrowth on its own ( 9% increase in wing area , Figure 1H , S ) , and synergized with Dachs over-expression , resulting in enhanced wing overgrowth ( Figure 1R , S ) . Together , these observations suggest that the functions of Zyx and Dachs in regulating growth are closely linked . However , the observation that Zyx depletion could enhance the small wing phenotype of a putative null allele of dachs ( Figures 1S , S1E , F ) [7] implies that Zyx also has some Dachs-independent influence on growth . Fat exerts a post-transcriptional influence on the levels of Wts protein [6] . The genetic placement of Zyx upstream of wts and within the Fat branch of the pathway suggested that Zyx might also affect Wts levels . Indeed , Zyx RNAi completely suppressed the reduction in Wts levels associated with fat RNAi ( Figures 5A , B , S2B ) . Thus , Zyx is genetically required for the mechanism that links Fat activity to the regulation of Wts protein levels . The influence of fat on Warts levels also requires dachs [6] . Zyx RNAi did not detectably affect Dachs localization ( Figure S4D , E ) , nor did Zyx RNAi affect Fat localization ( Figure S5E , F ) . In addition to its effects on Wts , fat mutation also decreases the levels of Ex at the sub-apical membrane [30]–[33] . Zyx RNAi was not able to reverse this effect of fat on Ex levels ( Figure S5G–N ) . Depletion of Zyx in the wing disc also did not have visible effects on F-actin ( Figure S5O , P ) . In addition to regulating transcription , Fat also regulates planar cell polarity ( PCP ) ( reviewed in [1] , [5] ) . PCP in the adult wing is manifest in the orientation of wing hairs , which point distally . The anterior , proximal wing is particularly sensitive to Fat-PCP signaling , and fat RNAi results in strong PCP phenotypes in this region , including reversals of hair polarity ( Figure S1M ) . PCP phenotypes have also been described in this region of dachs mutant wings [34] . Zyx RNAi , by contrast , had no detectable effect on wing PCP ( Figure S1N ) , and a PCP phenotype was also still detected in fat Zyx double RNAi wings ( Figure S1O ) . Genes previously identified as influencing Fat-PCP signaling ( i . e . , fat , ds , fj , app , dachs , lft ) also influence cross-vein spacing . Zyx RNAi wings sometimes have extra cross-veins , but by contrast to dachs mutants , the anterior and posterior cross-veins remain well-separated in Zyx RNAi flies ( Figure 1B , C ) , and the influence of fat on cross-vein spacing is not suppressed by Zyx ( Figure 1F ) . Our observations suggest that Zyx is specifically required for Fat-Hippo signaling , and not for Fat-PCP signaling , although because Zyx RNAi might not completely eliminate Zyx , we cannot exclude the possibility that low levels of Zyx are sufficient for PCP , but not for Hippo signaling . As our anti-Zyx sera did not work for immunostaining , we made use of a V5-tagged UAS transgene that rescues the Zyx RNAi phenotype ( Figure 1 ) to investigate the subcellular localization of Zyx in imaginal discs . We also examined a UAS-Ypet:Zyx transgene [35] . Although our localization studies are subject to the caveat that Zyx protein was over-expressed , the two different tagged Zyx proteins have similar localization profiles , and similar localization profiles were observed using different Gal4 drivers . Zyx was preferentially localized to the sub-apical membrane of disc cells ( Figure 6 ) . This sub-apical membrane staining was at the same apical-basal position as E-cadherin ( E-cad ) , and just basal to Fat ( Figure 6A–D ) . This is similar to the membrane localization of Dachs [7] . Indeed , when we compared Zyx and Dachs localization , using epitope-tagged constructs , we observed that the membrane staining is at the same apical-basal position and that they partially co-localize ( Figure 6G , H ) . A distinguishing feature of Dachs localization is its polarization within the plane of the epithelium , which occurs in response to the Fj and Ds gradients ( Figure 6J ) [7] , [9] . Zyx , by contrast , is not planar-polarized ( Figure 6I ) ; hence , Zyx and Dachs are expected to overlap on only one side of wing disc cells . A distinguishing feature of Zyx staining is that it often displays puncta of larger , more intense staining at the vertices where three cells meet ( Figure 6G ) . Intriguingly , Ex protein also displays uneven staining , but Ex puncta are partially complementary to Zyx puncta ( Figure 6E , F ) . These observations suggest that even though Ex and Zyx localize to a similar apical-basal position , they assemble into distinct protein complexes . Dachs localization was not visibly affected by RNAi of Zyx ( Figure S4E ) , nor was Zyx localization affected by mutation of dachs ( Figure S5B ) , which indicates that neither protein depends upon the other for its localization . Zyx localization was also not visibly affected by mutation or RNAi of fat , ex , or wts ( Figure S5 and unpublished data ) . The similar genetic requirements for Zyx and dachs in Fat-Hippo signaling , together with their partial co-localization in imaginal discs , raised the possibility that Zyx and Dachs might interact . This was investigated by expressing tagged isoforms in cultured Drosophila S2 cells and assaying for physical interactions through co-immunoprecipitation . Indeed , Zyx and Dachs could be specifically co-precipitated from S2 cells ( Figure 7B ) . This observation suggests that Dachs and Zyx can interact directly , although it is also possible that they interact indirectly through a larger complex including endogenously expressed proteins within S2 cells . As Dachs can also associate with Warts in co-immunoprecipitation assays [6] , and both Zyx and dachs are required for the fat-dependent regulation of Wts levels , we also investigated binding between Zyx and Wts . When tagged full-length proteins were co-expressed in S2 cells , little or no Zyx-Wts co-precipitation was detected ( Figure 7C , H ) . However , in addition to their role in Hippo signaling , functions for LATS proteins have also been identified in mitosis , and LATS1 has been localized to the mitotic apparatus [36] , [37] . In the context of a study of mitotic functions of LATS1 , it was reported that the C-terminus of human Zyxin , including the LIM domains , could bind to human LATS1 , even though full-length Zyxin did not bind [36] . When we expressed a C-terminal polypeptide comprising the LIM domains of Zyx ( Zyx-LD ) in S2 cells , only very low levels of protein could be detected ( Figure 7B–D ) . Nonetheless , this C-terminal polypeptide bound efficiently to Wts ( Figure 7C ) . Thus , the LIM domains of Zyx can associate with Wts , but this association is normally inhibited within full-length Zyx . The discovery of this latent ability of Zyx to bind Wts , together with our discovery of Zyx-Dachs binding , and previous identification of Dachs-Wts binding [6] , indicates that Dachs , Zyx , and Wts each have the ability to bind to one another . To gain further insight into complex formation among these proteins , we mapped their interaction domains . Wts bound to the LIM domains of Zyx . Dachs , by contrast , bound most strongly to the C-terminal LIM domains but also bound to the N-terminal half of Zyx ( Figure 7B ) . Dachs contains a large central myosin motor domain and could bind to both Zyx and Wts through this motor domain ( Figure 7D , G and unpublished data ) . Zyx-LD bound to Wts through a region N-terminal to the Wts kinase domain ( Figure 7E ) . Dachs bound both to this region and also to the Wts kinase domain ( Figure 7F ) . Thus , Zyx , Dachs , and Wts interact with each other through partially overlapping domains . To assay for potential sequential , cooperative , or competitive interactions amongst Zyx , Dachs , and Wts , we examined binding interactions when all three proteins were co-expressed together in S2 cells . A key feature of Zyx's interactions with Wts is that full-length Zyx does not bind efficiently to Wts , but the LIM domains do . However , we found that Dachs enhanced the co-precipitation of full-length Zyx with Wts ( Figure 7H ) . Two basic models for this stimulation of Zyx-Wts association by Dachs can be envisioned: ( a ) Dachs might bridge Wts and Zyx within a Wts-Dachs-Zyx complex , or ( b ) Dachs might trigger a conformational change in Zyx that reveals the latent Wts-binding activity of the Zyx LIM domains ( Figure 8A , B ) . By employing V5 epitope tags on both Zyx and Dachs , and assaying their co-precipitation with FLAG-tagged Wts , we could directly compare their association with Wts . A simple trimeric complex model ( e . g . , one subunit each of Zyx , Wts , and Dachs ) would predict that Zyx and Dachs should be present within the Wts trimeric complex at equal levels . However , we found instead that Zyx could be much more abundant in Wts complexes than Dachs ( Figure 7H ) . This suggests that rather than remaining stably associated with Zyx and Wts in a trimeric complex , Dachs is able to stimulate a conformational change in Zyx that exposes the LIM domains and enables them to bind Wts . Consistent with this model , Dachs stimulated Zyx binding to Wts but did not stimulate the binding of Zyx-LD to Wts ( Figure S6A ) . Zyx is a Drosophila member of a group of cytoskeletal-associated proteins with three C-terminal LIM domains [38] . These comprise two families: the Zyxin family , which in vertebrates includes Zyxin , Lipoma preferred partner ( LPP ) , and Thyroid-receptor interacting protein 6 ( TRIP6 ) , and the Ajuba family , which in vertebrates includes Ajuba , LIM domain containing 1 ( LIMD1 ) , and Wilms tumor protein 1-interacting protein ( WTIP ) . Drosophila have a single member of each family; Zyx is a member of the Zyxin family , and Ajuba LIM protein ( Jub ) is a member of the Ajuba family . Ajuba has been reported to interact with a human homologue of Warts , LATS2 [39] , and Das Thakur et al . ( 2010 ) recently reported that mutation or RNAi-mediated depletion of Jub reduces growth through interactions with the Hippo pathway , and through genetic and protein interaction experiments positioned Jub as a regulator of Wts [40] . In agreement with this , we found that RNAi-mediated depletion of Jub reduces wing growth ( Figure 1M , N , S ) , expression of Hippo pathway target genes , and nuclear Yki ( Figure S7 ) , and that wts is epistatic to Jub ( Figure 3C ) . As for Zyx , depletion of Jub did not detectably influence wing hair PCP ( Figure S1P , K ) . The determination that Zyx and Jub are each genetically required for Hippo signaling suggests that they have distinct functional roles , and consistent with this , we observed that over-expression of Zyx could not rescue Jub RNAi phenotype ( Figure S1H ) and that Zyx Jub double RNAi induced an even greater reduction of wing size than when they were expressed individually ( Figure 1O , S ) . Das Thakur et al . ( 2010 ) did not address the relationship of Jub to upstream regulators of Hippo signaling . Intriguingly , we found that depletion of Jub suppressed both fat and ex phenotypes . This suppression was evident upon examination of adult wings ( Figure 1G , K , S ) , expression of downstream target genes in wing discs ( Figures 3F , 4E , F ) , and the sub-cellular localization of Yki ( Figure 4I , L ) . Thus , by contrast to Zyx , which functions specifically within Fat-Hippo signaling , Jub is required for both Ex-Hippo and Fat-Hippo signaling . This observation confirms that these two LIM-domain proteins have functionally distinct roles within the Hippo pathway . The distinct genetic role of Jub in Hippo signaling is also reflected in distinct binding interactions . By contrast to the crucial role of Dachs in stimulating binding between full-length Zyx and Wts , full-length Jub binds efficiently to Wts , and full-length vertebrate homologues of Jub bind to LATS proteins [39] , [40] . Moreover , Jub bound only very weakly Dachs ( Figure S6B ) . Thus , although Zyx and Jub share the ability to associate with Wts through their LIM domains , both genetic and biochemical studies indicate that the regulation and consequences of these LIM-domain-Wts interactions are distinct .
Our characterization of Zyx identifies a role for it as a novel and integral component of the Hippo pathway , which is required for the Fat branch , but not the Ex branch , of Hippo signaling . Unlike most previously identified components , loss of Zyx reduces the activity of the key transcriptional effector of the pathway , Yki , and consequently its loss reduces organ growth . Genetic epistasis experiments position the requirement for Zyx in between fat and wts , and concordant protein binding experiments identify a Dachs-stimulated ability of Zyx to bind Wts protein . We infer that this association of Zyx with Wts then downregulates Wts , at least in part , by targeting it for degradation . Zyx localizes to the sub-apical membrane independently of Fat or Dachs . Since Fat regulates the localization of Dachs [7] , this regulated localization provides a mechanism by which Fat could modulate the interaction of Dachs with Zyx ( although we note that Fat might affect the activity of Dachs in addition to affecting its localization ) . Since Dachs stimulates Zyx-Wts binding , this regulated localization provides a means for Fat signaling to modulate Zyx-Wts binding . We infer that Dachs effects a conformational change in Zyx , as in the absence of Dachs a Zyx LIM-domains polypeptide binds efficiently to Wts , whereas full-length Zyx binds poorly . Intriguingly , the association of vertebrate homologues of Zyx and Warts can also be post-translationally regulated , as the ability of the LIM domains of human LATS1 to bind Zyxin is masked within full-length Zyxin , but uncovered by Cdc2-mediated phosphorylation , presumably due to conformational change [36] . We hypothesize that the ability of Dachs to bind to both the N-terminus and the LIM domains of Zyx enables it to effect a conformational change in Zyx , resulting in an open configuration that can bind to Wts ( Figure 8B ) . It is also possible that Dachs binding stimulates a post-translational modification of Zyx to induce a conformational change . Prior studies identified two mechanisms by which Fat signaling could influence Yki activity , as fat mutation reduces both the levels of Wts protein [6] and the amount of Ex at the sub-apical membrane [31]–[33] . It has not been possible to completely uncouple these two pathways for Fat-Hippo signaling , although the observation that over-expression of Wts can efficiently suppress fat overgrowth phenotypes , but only partially suppresses ex overgrowth phenotypes [30] , suggested that the influence of Fat on Wts levels might be more critical . Analysis of the influence of Zyx on Ex is complicated by its influence on ex transcription , but our observation that reduction of Zyx does not appear to suppress the influence of fat on Ex staining , even though it does suppress the influence of fat on Wts levels , also suggests that the influence of Fat on Wts levels might be more critical than its effects on Ex . Intriguingly , mutation of dachs did suppress the influence of fat on Ex levels [30] . Although it is possible that this difference between dachs and Zyx results from technical differences in the experimental paradigms ( e . g . , mutant clones versus RNAi ) , it is also possible that dachs can influence Ex levels independently from its association with Zyx . The discovery of the Fat-specific effect on Wts levels , by contrast to the Hippo-pathway-mediated effect on Wts kinase activity , established the concept of distinct mechanisms for regulating Wts—one that affects Wts levels and another that affects Wts activity [6] . Our identification of distinct genetic requirements for Zyx and Jub provide further support for this concept . As Jub is equally required for both Fat-Hippo and Ex-Hippo signaling and acts genetically between hippo and wts [40] , Jub appears to inhibit Wts activation . In our working model ( Figure 8C ) , the epistasis of Jub to fat could be explained by an increased activity of residual Wts , which then acts catalytically to repress Yki activity . Zyx is required for the influence of fat on Wts levels . We note that when measured within a whole tissue lysate , Wts levels are only reduced to approximately half their normal levels . However , as Wts appears to function within multi-protein complexes , including some components that can localize preferentially to the sub-apical membrane [41] , [42] , we hypothesize that Fat signaling affects a discrete pool of Wts within a complex at the membrane that is crucial for Hippo signaling , whereas there might be additional pools of Wts within the cell that are unaffected . We also note that while we clearly see effects on Wts protein levels , our results do not exclude the possibility that Fat signaling also influences Wts activity . Our characterization of Zyx and Jub also provides new tools for analyzing critical steps in Hippo signaling . For example , in addition to influencing Hpo and Wts kinase activity , it has been observed that Ex can bind directly to Yki and that when Ex is over-expressed it can repress Yki through a mechanism that involves direct sequestration of Yki , rather than regulation of Yki phosphorylation [43] , [44] . Because this direct repression mechanism was based on over-expression experiments , the extent to which it contributes to normal Yki regulation in vivo remained uncertain . The observations that Jub acts genetically upstream of wts , yet is required for ex phenotypes , suggests that Ex regulates Yki principally through its effects on Wts activity , rather than through direct interaction with Yki . The ability of Zyx LIM domains to interact with Wts is conserved in their human homologues [36] . Although the functional significance of this interaction in vertebrates has not yet been established , our observations raise the possibility that the oncogenic effects of human LPP mutations [13] could be due to an ability of these aberrant LPP fusion proteins to negatively regulate LATS proteins , resulting in inappropriate activation of YAP or TAZ . One of the most intriguing aspects of Zyxin family proteins is their role in mediating effects of mechanical force on cell behavior [18] . Zyxin family proteins can localize to focal adhesions of cultured fibroblasts , and this localization is modulated by mechanical tension [15] , [18] , [45] . The observation that increasing tension on stress fibers stimulates Zyxin accumulation at focal adhesions is intriguing in light of our observation that Zyx tends to accumulate at higher levels at intercellular vertices in imaginal discs , as these could be points of increased tension . As the association of unconventional myosins with F-actin can also be influenced by external force [46] , our discovery of binding between a myosin protein ( Dachs ) and Zyx raises the possibility that other myosins might also interact with Zyxin family proteins , which could potentially influence either their tension-based recruitment or their activity . Finally , we note that theoretical models of growth control in developing tissues have proposed that growth should be controlled by mechanical tension [47] , [48] , and direct evidence for mechanical effects on growth has been obtained in cultured cell models [49] . However , a mechanism for how this might be achieved has been lacking . Our discovery that Zyx , a member of a family of proteins implicated in responding to and transducing the effects of mechanical tension , is also a component of the Hippo signaling pathway , a crucial regulator of growth from Drosophila to humans , raises the intriguing possibility that Zyxin family proteins might form part of a molecular link between mechanical tension and the control of growth .
RNAi screening was conducted using lines from the NIG-Fly Stock Center ( http://www . shigen . nig . ac . jp/fly/nigfly/index . jsp ) , which were crossed to vg-Gal4 UAS-dcr2 or pnr-Gal4 UAS-dcr2 . Those with growth phenotypes were then re-screened for effects on Diap1 and Wg expression in imaginal discs by crossing to ci-Gal4 UAS-dcr2 or en-Gal4 UAS-dcr2 . All crosses were carried out at 28 . 5 C to obtain stronger phenotypes . Approximately 1 , 200 lines were examined in the initial screen ( Table S1 ) . Additional RNAi lines employed include ds [vdrc 36219] , fat [vdrc 9396] , d [vdrc 12555] , ex [vdrc 22994] , Zyx [NIG-32018R3] , Zyx [vdrc 21610] , wts [vdrc 9928] , wts [NIG-12072R1] , mats [vdrc 108080] , hpo [vdrc 104169] , Jub [vdrc 101993] , and Jub [vdrc 38442] . The effectiveness of fat and ex RNAi is illustrated in Figure S3E , F . Both Zyx RNAi lines gave similar effects on growth and gene expression in combination with multiple Gal4 lines and also behaved similarly in epistasis tests . UAS lines employed include UAS-dco3[29] , [48] , UAS-d:V5[9F] and UAS-d:V5[50] [7] , UAS-d:citrine[28] ( B . K . Staley , unpublished ) , UAS-Zyx:V5 , and UAS-Ypet:Zyx [35] . Gal4 lines employed include Dll-Gal4 , ex-lacZ en-Gal4 UAS-GFP/CyO;UAS-dcr2/TM6b , en-Gal4/CyO; th-lacZ UAS-dcr2/TM6b , ci-Gal4 UAS-dcr2[3]/TM6b , w UAS-dcr2[X]; nub-Gal4[ac-62] , w; AyGal4 UAS-GFP/C yO;UAS-dcr2/TM6b , y w hs-FLP[122]; AyGal4 UAS-GFP/CyO , tub-Gal80ts/CyO , Act-GFP; tub-Gal4 UAS-dcr2/ TM6b , w; tub-Gal4/CyO-GFP . MARCM clones were made by crossing y w hs-FLP[122] tub-Gal4 UAS-GFP/FM7 ; tub-Gal80 FRT40A/CyO to fat8 FRT40A/CyO , exel FRT40A/CyO , dGC13 FRT40A/CyO or y+ FRT40A ( as a control ) and UAS-zyxin:V5 . Flp-out clones were made by crossing y w hs-FLP[122]; AyGal4 UAS-GFP to UAS-zyxin:V5 or crossing AyGal4; UAS-d:citrine to y w hs-FLP[122]; UAS-zyxin:V5 . Genetic interaction of Zyx and dachs was examined by recombining nub-Gal4 with dGC13 and crossing to dGC13; RNAi-Zyx32018 . Adult wing phenotypes were scored by crossing UAS-dcr2; nub-Gal4 females to males of RNAi lines or Oregon-R males as a control . Wings of male progeny were photographed , all at the same magnification . For quantitation , between 9 and 12 wings per genotype were traced using NIH Image J , and wing areas were normalized to the average area in control males . Standard error of the mean ( s . e . m . ) and t tests were calculated using Graphpad Prism software . For analysis of gene expression in imaginal discs , ex-LacZ en-Gal4 UAS-GFP; UAS-dcr2 females were crossed to RNAi line males , and larvae were kept at 28 . 5 C until dissection . For analysis of Zyx:V5 or Ypet:Zyx localization , expression was driven by en-Gal4 , AyGal4 , or tub-Gal4 . Discs were fixed in 4% paraformaldehyde and stained using as primary antibodies: goat anti-ß-galactosidase ( 1∶1 , 000 , Biogenesis ) , mouse anti-Diap1 ( 1∶200 , B . Hay ) , rat anti-E-cad ( 1∶200 , DSHB ) , guinea pig anti-Ex ( 1∶2000 , R . Fehon ) , rat anti-Fat ( 1∶400 ) [29] , mouse anti-V5 ( 1∶400 , Invitrogen ) , mouse anti-Wg ( 1∶400 , DSHB ) , and rabbit anti-Yki ( 1∶400 ) [22] . F-actin was stained using Alexa Fluor 546 phalloidin ( 1∶100 , Invitrogen ) , and DNA was stained using Hoechst ( Invitrogen ) . Details of plasmid construction are in Text S1 . Co-immunoprecipitation assays were performed as described previously [6] . Cell lysates were cleared using protein G beads ( Sigma ) . Anti-V5 or anti-FLAG M2 beads ( Sigma ) were incubated with cell lysates overnight at 4°C , then washed six times with RIPA buffer and boiled in SDS-PAGE loading buffer . Primary antibodies used for blotting include rabbit anti-V5 ( 1;10 , 000 , Bethyl ) , mouse anti-V5 ( 1∶10 , 000 , Invitrogen ) , and mouse anti-FLAG M2 ( 1∶10 , 000 , Sigma ) , and were detected using anti-mouse IRdye680 and goat anti-rabbit IRdye800 ( 1∶10 , 000 , LiCor ) and scanning on a LiCor Odyssey . For analysis of Wts protein levels , tub–Gal4 UASdcr2/ TM6b females were crossed to white ( control ) , RNAi-fat , RNAi-Zyx , RNAi-fat; RNAi-Zyx , or UAS-Zyx:V5 males , and wing discs were dissected from third instar larval progeny and lysed in RIPA buffer . Amounts loaded were adjusted to try to load equivalent amounts of total protein in each lane . Wts was detected using a published Wts anti-sera [6] at 1∶4 , 000 . Protein bands were detected using anti-mouse IRdye680 and goat anti-rabbit IRdye800 ( 1∶10 , 000 , LiCor ) and scanning on a LiCor Odyssey . Bands were quantified using LiCor Odyssey software . Relative Wts levels were determined by comparison to bands detected by anti-Actin antibodies ( mouse anti-Actin at 1∶5 , 000 , Calbiochem ) . To enable the relative levels of Wts to be averaged across different blots , we normalized the ratios on each blot to that detected for the control lane , which was set as 1 . For confirmation of the influence of Zyx RNAi on Zyx protein levels , tub–Gal4 UASdcr2/TM6b females were crossed to white ( control ) , or RNAi-Zyx32018 , and cultured at 29 C , and wing discs were dissected from third instar larval progeny and lysed in RIPA buffer . A rabbit anti-Zyx sera was used at a 1∶2 , 000 dilution , and subsequently the blot was re-probed with rabbit anti-actin ( 1∶10 , 000 , Sigma ) . Fluorescent detection was performed as described above . Anti-Zyx sera was obtained by immunization of rabbits with a KLH conjugated peptide ( KRRLDIPPKPPIKY ) , performed by Open Biosystems . | Processes that control cell numbers are essential during normal development , when they are required to generate organs of the correct size , and during cancinogenesis , when they influence tumor growth . The Hippo pathway is an intercellular signaling pathway that relays information about cell-cell contact and cell polarity to a signal transduction pathway that regulates the transcription of genes controlling cell numbers . The role of Hippo signaling in controlling growth is conserved from fruit flies to humans , but many aspects of the Hippo signal transduction pathway remain poorly understood . In this article , we identify Zyx as a previously unknown component of the Hippo pathway in Drosophila , and characterize its role within the pathway . We show that Zyx plays an essential role in a branch of Hippo signaling that involves the transmembrane receptor protein Fat and its target Dachs , which is a myosin family protein . Our results suggest a model in which Fat regulates the localization of Dachs , Dachs subsequently binds Zyx , stimulating its binding with the kinase Warts/Lats , and thereby regulates downstream signaling events . Zyx is conserved in vertebrates and we suggest that vertebrate Zyx proteins might also be involved in the regulation of Hippo signaling and , thereby , organ growth . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"developmental",
"biology/organogenesis",
"cell",
"biology/cell",
"growth",
"and",
"division",
"cell",
"biology/cell",
"signaling"
] | 2011 | Zyxin Links Fat Signaling to the Hippo Pathway |
We present a rigorous statistical model that infers the structure of P . falciparum mixtures—including the number of strains present , their proportion within the samples , and the amount of unexplained mixture—using whole genome sequence ( WGS ) data . Applied to simulation data , artificial laboratory mixtures , and field samples , the model provides reasonable inference with as few as 10 reads or 50 SNPs and works efficiently even with much larger data sets . Source code and example data for the model are provided in an open source fashion . We discuss the possible uses of this model as a window into within-host selection for clinical and epidemiological studies .
The protozoan parasite Plasmodium falciparum ( Pf ) is the cause of the vast majority of fatal malaria cases , killing at least half a million people a year [1–3] . The parasite’s ability to develop resistance to drugs and the rapid spread of that resistance across geographically-separated populations presents a constant threat to international control efforts [4–6] . While research has elucidated many genetic factors this process , much of the genetic epidemiology of the parasite—including the effective recombination rate and the rate of gene flow across populations—is still unclear [5 , 7 , 8] . Understanding the implications of multiplicity of infection ( MOI ) , where multiple strains appear to be present within a single patient’s bloodstream , is a long-standing challenge [9–13] . While MOI-focused studies implicate MOI levels with a range of conditions , including clinical severity [14] , age-specific severity [15–18] , parasitemia levels during pregnancy [19] , and other effects [20–23] , there is no broad consensus about its role in controlling the course of an infection . Still , a wide variety of studies and genetic assays—most commonly through typing the MSP genes—show MOI as a regular feature of clinical Pf isolates [24–26] . WGS technologies applied to Pf extracted directly from infected patients’ bloodstreams provide an unprecedented window into the structure of genetic mixture within samples [27 , 28] . Initial work on this data shifted focus from estimating MOI to analysis based on inbreeding coefficients [13 , 29–31] . These metrics , a form of F-statistic , give an estimate of the departure of within-sample allele frequencies from those expected under a Hardy-Weinberg-type equilibrium with the ambient population . From this perspective , each patient’s bloodstream is seen as a subpopulation comprised of an admixture of all strains in the local environment , ranging from a perfectly random sampling of all nearby strains ( panmixia ) to the repeated sampling of just a single strain ( unmixed ) . The initial study applying WGS to clinical Pf isolates from eight countries on three continents showed the parasite to exhibit significant population structure at continental scales , with the amount of subpopulation structure varying significantly among regions [27] . Employing an F-statistic approach to measure the inbreeding coefficient from thousands of single nucleotide polymorphisms ( SNPs ) , this work also argued that the degree of mixture varies significantly across populations , with highly mixed samples occurring relatively frequently in west Africa but only occasionally in Papua New Guinea . The authors suggest an association between increased levels of observed mixture and increased transmission intensity in the local environment . Transmission intensity , the rate at which individuals are infected with Pf , likely determines some part of the frequency of out-crossing within parasite populations and so may be critical to understanding gene flow and strategies for resistance control [32] . In this paper , we present a statistically rigorous model that synthesizes these two distinct and previously disparate approaches to analyzing Pf clinical mixtures: assessing the number of distinct genetic types within a sample ( the MOI approach [31] ) and measuring the degree of panmixia with respect to the local population ( the panmixia approach [33] ) . The model makes two significant innovations: first , a reversible jump Markov Chain Monte Carlo ( MCMC ) implementation to capture uncertainty in the number of strains , and second the inclusion of a panmixia term to deal with unexplained variation in the mixture . This work possesses similarities in character to the COIL algorithm [34] , but can capture more complex mixture structure and is geared toward analyzing WGS data ( >1000 SNPs ) rather than a small number of SNPs ( ∼50 SNPs ) . This model centers around how the two sub-models—MOI and panmixia—contribute to the observed within-sample non-reference allele frequencies ( WSAF ) as they relate to the population-level non-reference allele frequencies ( PLAF ) . For clarity , we will deprecate the use of non-reference in front of the term allele frequency , since they are all calibrated in this fashion . We will use the acronyms WSAF to denote the within-sample allele frequency and PLAF to denote population-level allele frequency to avoid confusion about the particularly allele frequency being indicated . The goal of the model is to explain observed ‘bands’ that emerge when examining SNPs WSAF as a function of their PLAF ( Fig 1 ) . The model assumes ( 1 ) that the number of bands is a consequence of the number of distinct strains present within a sample , ( 2 ) that SNPs are unlinked , and ( 3 ) that unexplained variation is assumed to be due to a small fraction of genomes sampled under panmixia . To distinguish from an inbreeding coefficient—a similar but distinct concept—we refer to this fraction as a panmixia coefficient . The collection of WSAF bands then appears as a function of the finite mixture of the strains , with the slope in WSAF bands with respect to the PLAF explained by both the SNP distribution and the panmixia coefficient . Fig 2 lays out how the consequent banding patterns can arise . In the simplest case , a sample is composed of a single , unmixed strain , and all SNPs exhibit a WSAF of zero or one ( see Fig 2 ( a ) ) , based on whether they agree with the reference . Consequently , the WSAF is independent of PLAF , leading to two flat bands at these values . We call these samples unmixed , since this is how a single strain with some divergence from the reference will appear . In the case where there are a finite number of strains mixed within a sample , then at a given SNP position some number of the strains will exhibit a reference allele and some a non-reference allele . The WSAF for that SNP is determined by the proportions of non-reference strains in the sample mixture . Observing many SNPs displays ‘bands’ of constant WSAF across the PLAF . Thus , for K component strains there are 2K possible combinations of biallelic states , leading to that number of bands . A fraction of the Pf organisms present within the blood may not be from any of the dominant strains . We model these as randomly sampled from the local population according to simple panmixia . Observationally , this leads to a consistent change in the slope of each of the bands . To see this , consider an admixture of two distinct Pf populations: a single strain , representing 1 − α of the within-sample genomes , and the remaining α that we assume follow panmixia . The α tilt in the WSAF arises from the fact that for this proportion of organisms the probability of sampling non-reference allele is proportional to the PLAF ( Figs 1 ( c ) and 2 ( c ) ) . Samples with high K appear to have additional tilt due to the higher probability of non-reference alleles occurring at high PLAF ( Figs 1 ( d ) and 2 ( d ) ) . The paper proceeds as follows . We first detail the structure of the WGS data , introduce some notation , and the essential mathematical structure of the model . We then present an extensive simulation study on the performance of the model , an application of the model to artificial laboratory mixtures , and an examination of its application to field isolates collected from northern Ghana . We conclude by discussing the strengths and weaknesses of the model , a means of experimental validation , and potential consequences for the etiology of clinical malaria .
The field WGS data come from Illumina HiSeq sequencing applied to Pf extracted from 419 clinical blood samples collected from infected patients in the Kassena-Nankana district ( KND ) region of Upper East Region of northern Ghana . Collection occurred over approximately 2 years , from June 2009-June 2011 . The raw sequence reads for these data are accessible through the PF3K project https://www . malariagen . net/projects/parasite/pf3k . This includes data from the MalariaGEN Plasmodium falciparum Community Project on www . malariagen . net/projects/parasite/pf . On the website for this method , we provide read count data subsampled from the full data set . The artificial laboratory samples were sequenced and called per protocols given in [35] . The raw sequence data is available through the European Nucleotide Archive with the accessions available in the S1 Text . The full sequencing protocol and collection regime are described in [27] . After quality control measures , all samples were examined , and following a documented protocol comparing against world-wide variation , 198 , 181 single-nucleotide polymorphisms ( SNPs ) were called [27] . These are exclusively coding SNPs found outside of the telomeric and subtelomeric regions that exhibit unusual structural properties . Each SNP xcall provides the number of reference and non-reference read counts observed at each variant position within the genome , ascertained against the the 3D7 reference [36] . These data were exhaustively examined for spurious heterozygosity and evidence of DNA contamination , with mixed calls verified using time-of-flight mass spectrometry at greater than 99% accuracy [27] . For this project , we further filtered the data . First , multiallelic positions were reclassed as biallelic . We then excluded positions that exhibited no variation within the KND samples , any level of missingness ( no read counts observed ) , or minor allele frequency less than 0 . 01 . To remove low quality samples , we removed those with more than 4 , 000 SNPs missing and fewer than 20 read counts , following an inflection point observed in S1 Fig . These cleaning measures left 2 , 429 SNPs in 168 samples . These SNPs exhibit desirable properties for model inference—high and consistent coverage across all samples—that could likely be expanded to non-coding or less stringent cleaning standards without issue . More than 95% of the remaining samples’ sequencing was completed without PCR amplification . We observe little apparent population structure among the samples , evidenced either by principal components analysis or a neighbor-joining tree of the pairwise difference among samples ( S2 Fig ) . The data preparation scripts are available with the source code for the model , https://github . com/jacobian1980/pfmix/ . Following the data preparation and cleaning , our analysis begins with a set of N = 168 clinical samples , each composed of M = 2 , 429 SNPs . At each SNP j within each clinical sample i , we observe rij reads that agree with the reference genome and nij reads that do not agree . The total number of read counts in sample i at SNP j is then nij + rij . For a sample i , we write the complete data across all SNPs as D i = [ ( r i 1 , n i 1 ) , ⋯ , ( r i M , n i M ) ] . For each SNP j , we associate a PLAF pj . The collection of all pj we refer to as P . Conditional upon the number of strains K , there are 2K bands , indexed by r = 1 , ⋯ , 2K . The full collection of bands we call Q , with qijr indicating the WSAF for sample i at SNP j in band r . The probability of a SNP lying within the distinct bands across the PLAF is specified by a mixture component λr , which is a function of the PLAF detailed below . The degree of panmixia in a sample is given by α , a value between zero and one . A complete list of the model parameters is given in Table 1 . Statistically , the model takes the form of a finite mixture model with the mixture components associated with individual bands [37 , 38] . We take a Bayesian approach to inference and construct the model by giving an overall rationale for the decomposition of the posterior distribution , and then justify the appropriate choice of probability distributions for each of the terms [39] . We infer the model parameters using a standard reversible jump MCMC approach [40 , 41] with one exception: we first calculate maximum-likelihood estimates ( MLE ) for P across all samples and then treat these as fixed when inferring the remaining parameters [42] . This choice is motivated by statistical expedience and computational speed: except for P , the parameters of the model are independent across samples and so this approximation enables the algorithm to infer parameters in parallel rather than jointly . This avoids the difficulties of performing inference on the number of strains within all samples simultaneously . Running in parallel also increases the computational speed of the implementation by at least an order of magnitude . Since the sample collection is large enough that the PLAF is nearly independent of any given sample , we do not expect this approximation to significantly bias inference . For each SNP j , the MLE derives from treating the non- and reference reads within a sample as coming from a binomial distribution with parameter pj . This leads to: p ^ j = ∑ i N n i j / ∑ i N ( n i j + r i j ) . To infer the number of strains , K , for each sample , we employ a pair of complementary split/merge reversible jump MCMC moves . To specify the split move first not that in moving from K → K + 1 that the transformation only affects the parameter W . If we randomly select wk , 1 ≤ k ≤ K , then we can split this into two components , u ⋅ wk and ( 1 − u ) ⋅ wk , where u is drawn from a uniform distribution . This establishes a diffeomorphism between parameters at K and K + 1 with Jacobian wk . The proposal ratio is ( K2 − K ) /K = K − 1 . The acceptance ratio then is the product of the proposal ratio , Jacobian , the likelihood ratio , and the prior likelihood . The merge move randomly selects two states , k1 and k2 , and merges them to k′ by setting w′ = wk1 + wk2 . The Jacobian and proposals are the reciprocal of those for the split move . Conditional on P and K , for each of the three parameters , α , W , and ν , we propose new values directly from the prior distribution . This leads to Metropolis-Hastings ratios almost solely dependent on the ratio between the likelihood and priors for the proposed state to those for the current . The inference scheme is implemented in set of scripts for the R computing language , and can be found under the Academic Free License at https://github . com/jacobian1980/pfmix/s . For a single sample with K = 5 , a sufficiently long MCMC run takes approximately 10 minutes on a single high-performance computing core .
To demonstrate the efficacy of the model and our implementation , we present a simulation study examining the algorithm’s performance under a range of simulated data . We consider two distinct aspects of the inference: how well the model infers the number of strains , and , conditional upon that number , how well it infers the model’s other parameters . We simulate data from the model in the following way . Conditional upon the number of SNPs ( M ) , panmixture coefficient ( α ) , number of strains ( K ) and the sum of the read counts ( C ) we draw a vector of probabilities ( W ) from a uniform Dirichlet distribution . We combine the values of W in all possible permutations to create the 2K bands and assign the PLAF for the SNPs evenly from 1/M to 1 , so that the j th SNP has PLAF j M . For each SNP , we first probabilistically select the band it occupies according to Eq ( 6 ) . We then simulate read counts from the likelihood ( Eq 5 ) with qijr per Eq ( 8 ) . For all simulations , we set ν = 10 . We run the simulation across the range of values for M , α , K and C given in Table 2 . For each parameter set , we create 10 independent realizations . We apply the algorithm to 18 artificial laboratory mixtures . These artificial samples were generated by taking stock of two standard Pf lines , DD2 and 7G8 , and adding them together in the fixed proportions given in S1 Table , and completing then Illumina sequencing and variant-calling with using the same protocols as [27] . Samples had a median of 65 reads for the variants considered here . Complete sequencing protocols and laboratory methods detailed in [35] ( data available at European Nucleotide Archive ) . Both strains have high-confidence reference sequences . We subsample 2000 SNPs from the 23 , 109 SNPs available for comparison based on non-reference WSAF . The results in S1 Table show very strong agreement between the laboratory and inferred mixtures . The inferred α for all samples was less than 0 . 001 and had Bayes factor for non-zero α as less than 1 , indicating that the samples have little unexplained mixture observed relative to the field samples . Applying the algorithm to the 168 high-quality samples from KND , we observe numbers of strains ranging from 1 to 7 , with α falling between 0 and 0 . 14 , and a moderate correlation between K and α ( Fig 5 ) . The largest subset of samples were unmixed , with K = 1 and α < 0 . 01 , though the majority of samples exhibit low but noticeable levels of admixture , with K = 2 , 3 , 4 and 0 . 01 ≤ α ≤ 0 . 03 . A small number of samples exhibit complex mixtures , with K > 4 and α typically greater than 0 . 02 . These samples also exhibit the most variance in the posterior estimate of K , frequently ranging from 3 to 8 . To examine the necessity of the panmixia model to capture unexplained variation in the field samples , we calculate a Bayes factor for each sample under the two models , M0: α = 0 and M1: α ≠ 0 . Since this is a single parameter , we employ the Savage-Dickey ratio calculation as in [43] . We find that 78 samples give factors larger than 10 , indicating strong evidence for M1 , and 9 samples give factors larger than 100 , indicating overwhelming evidence for M1 . To visually inspect the quality of the results , we generate figures for each of the samples showing the observed WSAF and PLAF data , the inferred model structure , and data simulated under the inferred model following the observed PLAF . We show examples of these plots for three typical samples in Fig 6 . Nearly all samples ( 158/168 ) , across all different mixture patterns , show strong visual correspondence between the observed and model-simulated data . Samples where PCR amplification was used ( 9 samples ) exhibit no unusual features other than low values for α relative to the remaining samples . We also observe a strong correlation between the inferred number of components and an estimate for the inbreeding coefficient for each sample ( Fig 7 ) [29] . These results are consistent with the high rate of MOI previously observed in Ghanaian clinicial samples [24 , 44 , 45] .
In this work we show how to infer strain mixture within Pf isolates using WGS with two improvements over previous efforts: an additional model for unexplained variation based on a panmixia and a reversible jump implementation that accounts for uncertainty in the underlying number of strains . Simulations show that the model can perform accurate inference ( MSE < 0 . 05 for strain proportions ) with as few as 50 SNPs and 10 read counts per SNP . Simulations with more than 100 SNPs or at least 25 read counts give highly accurate results ( MSE < 0 . 02 ) . In artificial laboratory mixtures the model provides excellent agreement with baseline mixture . In field samples the model provides strong agreement with observed data and evidence based on Bayes factors indicates that some unexplained variation is present in a significant fraction of samples . While the method works efficiently in practice , a number of possible improvements could strengthen its statistical performance . Most immediately , creating a full Bayesian approach rather than the parallelizing implementation here—while likely not improving the parametric inference for individual samples—would provide the full posterior distribution across all samples . The panmixia model is one of several possible approaches to dealing with additional within-sample variation that rigorous model comparison could reveal . The model also does not perform haplotype phasing to resolve the sequence of the underlying strains [46–48] . The analysis here suggests that a method for estimating haplotypes would be straight-forward for some samples but difficult for others ( say , when α is greater than 0 . 05 ) . Researchers may be particularly interested in whether , in these phased samples , particular stretches of the genome appear more or less frequently in the dominant strains than others , indicating structures of immunological or environmental selection . This is a natural avenue for statistical methods development . The model makes a number of simplifying assumptions that may be violated in practice . The model presumes that SNPs are unlinked and consequently independent for the purpose of calculating the likelihood . Given the high recombination rate of Pf this assumption may hold for the majority of pairs of SNPs , but neglects correlations that appear locally ( ∼ 10 kB ) . However , we expect that this independence assumption serves to moderately weaken the inferential power of the model rather than cause any type of bias since it effectively fails to include possibly informative data . More problematic is the model’s implicit assumption of limited population structure . In the case of the KND samples , and perhaps in much of west Africa , this assumption appears supported [27 , 49] . In other contexts , specifically southeast Asia , recent population bottlenecks and selection suggest that this assumption will be violated [50] . The consequences on this model inference are unknown but may be partially resolved with appropriate simulation studies . The model will work with any technology capable of typing multiple variants and where the measurement of the fraction of non-reference variants is unbiased . It was developed for WGS data but is not specific to the sequencing employed and should work similarly for Illumina , 454 and Pacific Bioscience read technologies . As noted in the results , we observe that the small number of field samples where PCR amplification was used did not appear unusual other than exhibiting relatively low α values . This is could be due to preferential amplification of the dominant strains , suggesting that PCR-based approaches may obscure some aspects of natural infections . This model is not appropriate for data from RFLP assays or DNA microarrays without substantial modification . In principle , the model can be explicitly tested against experimental mixtures more complex than those presented above . Laboratory facilities with the capacity to store many field strains ( >100 ) could generate artificial samples in an experimental analog of our simulation procedure . Starting with N unmixed strains at equal dilution , they could create mixtures by first fixing the required sequencing volume as η , and the desired parameters for panmixia ( α ) , number of component strains ( K ) , and their mixture parameters , W . For the finite mixture component , they would then combine volumes of η · W from the K strains . For the panmixture component , they would then fix some large number but experimentally feasible number of strains ( say 50 ) and randomly sample from all of them a volume of η/50 . Combining these into a final sample and applying WGS sequencing , will yield data that we hypothesize will closely follow the integrated model outlined above , with ν capturing the experimental variation . Naturally , consistent results would indicate the sufficiency of the model but not its necessity , holding out the possibility of a more minimal description . These results could be further compared against other next-generation technologies—such as single-cell sequencing—that have been deployed to understand Pf clinical mixtures [51] . The model presents an important new tool for interrogating the biology of clinical Pf infections . The ability to measure the structure of strain mixture connects to many aspects of Pf epidemiology including seasonality , transmission intensity , outcrossing , and rates of gene flow . It also presents a means for clarifying the poorly detailed structure of intra-host infection dynamics , such as strain selection or density-dependent selection [52] , by resolving how the model parameters change within the course of an infection or in response to drug intervention . This approach can serve as a means for researchers to empirically resolve these hypotheses . | Since the 1960’s researchers have observed that Plasmodium falciparum infections , the cause of most severe malaria , are frequently composed of several different strains of the parasite . In this work , the authors use Bayesian methods on whole genome sequence data to model the structure of these mixtures . Results from this method are consistent with previous approaches but provide finer resolution of these mixtures . As whole genome data in malaria research becomes increasingly common , this work will allow researchers to delve further into the within-host dynamics of the parasite , a much-discussed but previously difficult-to-access aspect of this disease . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"sequencing",
"techniques",
"medicine",
"and",
"health",
"sciences",
"plasmodium",
"falciparum",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"simulation",
"and",
"modeling",
"probability",
"distribution",
"mathematics",
"protozoans",
"molecular",
"biology",
"techniques",
"research",
"and",
"analysis",
"methods",
"malarial",
"parasites",
"molecular",
"biology",
"inbreeding",
"probability",
"theory",
"nucleotide",
"sequencing",
"heredity",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"genomics",
"organisms"
] | 2016 | Inferring Strain Mixture within Clinical Plasmodium falciparum Isolates from Genomic Sequence Data |
Precise positioning of the cell division site is essential for the correct segregation of the genetic material into the two daughter cells . In the bacterium Myxococcus xanthus , the proteins PomX and PomY form a cluster on the chromosome that performs a biased random walk to midcell and positively regulates cell division there . PomZ , an ATPase , is necessary for tethering of the cluster to the nucleoid and regulates its movement towards midcell . It has remained unclear how the cluster dynamics change when the biochemical parameters , such as the attachment rates of PomZ dimers to the nucleoid and the cluster , the ATP hydrolysis rate of PomZ or the mobility of PomZ interacting with the nucleoid and cluster , are varied . To answer these questions , we investigate a one-dimensional model that includes the nucleoid , the Pom cluster and PomZ proteins . We find that a mechanism based on the diffusive PomZ fluxes on the nucleoid into the cluster can explain the latter’s midnucleoid localization for a broad parameter range . Furthermore , there is an ATP hydrolysis rate that minimizes the time the cluster needs to reach midnucleoid . If the dynamics of PomZ on the nucleoid is slow relative to the cluster’s velocity , we observe oscillatory cluster movements around midnucleoid . To understand midnucleoid localization , we developed a semi-analytical approach that dissects the net movement of the cluster into its components: the difference in PomZ fluxes into the cluster from either side , the force exerted by a single PomZ dimer on the cluster and the effective friction coefficient of the cluster . Importantly , we predict that the Pom cluster oscillates around midnucleoid if the diffusivity of PomZ on the nucleoid is reduced . A similar approach to that applied here may also prove useful for cargo localization in ParABS systems .
The formation of protein patterns and the intracellular positioning of proteins is a major prerequisite for many important processes in bacterial cells , such as cell division . In order to maintain the genetic content of the bacterial cell , the chromosome ( nucleoid ) is duplicated during the cell cycle and must be segregated into the two cell halves prior to cell division . The future division site is defined by the FtsZ ring , which forms at midcell and recruits the cytokinetic machinery . Interestingly , FtsZ is highly conserved in bacteria , while the protein systems responsible for the positioning of the FtsZ ring , and with it the cell division site , are not [1–3] . Recently , Schumacher et al . identified a set of proteins , called PomX , PomY and PomZ , in Myxococcus xanthus cells that are important for midcell localization and formation of the FtsZ ring [4–6] . PomZ is an ATPase , which belongs to the family of ParA / MinD ATPases [3] . It binds non-specifically to DNA in its dimeric , ATP-bound state , and its activity is stimulated by interactions with PomX , PomY and DNA . PomX and PomY form a single cluster , which is tethered to the nucleoid via PomZ dimers bound to the chromosome . Starting from an off-center position near one nucleoid pole , the cluster moves towards midnucleoid , coinciding with midcell [5] . When the cluster has reached midcell , the FtsZ ring forms there and the cell divides . During cell division , the cluster splits into two halves , such that each half is located at one pole of the nucleoids in the daughter cells , and the same cycle repeats . Notably , the Pom proteins localize to midcell before FtsZ and also in the absence of FtsZ [4 , 5] . The mechanism underlying midcell localization of the FtsZ ring is well understood in Escherichia coli cells [7–21] . Here , Min proteins ( MinC , MinD and MinE ) guide the formation of the FtsZ ring at midcell . Importantly , the system for midcell localization in E . coli and M . xanthus differ substantially , even though both systems contain an ATPase ( PomZ and MinD , respectively ) and perform the same task in the cell , i . e . midcell sensing . First , the scaffold to which the ATP-bound ATPase binds is different: MinD binds to the cell membrane and PomZ to the bacterial nucleoid in the cytoplasm . Second , MinD-bound MinC inhibits [22] , whereas the Pom cluster promotes FtsZ ring formation at midcell [5] . Finally , the observed protein patterns differ: the Pom proteins colocalize in a cluster that moves towards midcell , while the Min proteins , which do not form a cluster , oscillate from pole to pole [22 , 23] . Conversely , the Pom system is mechanistically more similar to plasmid and chromosome segregation systems that involve a ParABS system . Like the Pom system , plasmid and chromosome segregation systems make use of an ATPase that shuttles one or several cargoes ( such as a plasmid , a partition complex or a protein cluster ) along the nucleoid . To ensure the equal distribution of low-copy number plasmids to the daughter cells , they are tethered to the nucleoid and positioned at equal distances along the nucleoid by ParABS systems [24–26] . A ParABS system consists of the proteins ParA and ParB , and a DNA sequence , parS . ParA proteins are ATPases , which bind non-specifically to DNA as ATP-bound dimers [27–29] . Their ATPase activity is stimulated in the presence of ParB [30–32] , which binds to the parS sequence on the chromosome ( to form the partition complex ) or on the plasmid [3] . Besides plasmid and chromosome segregation [33] , ParABS systems are also important for the positioning of cellular components ( e . g . chemotactic clusters or carboxysomes ) [34 , 35] . Several different cargo dynamics involving ParABS systems have been observed . For one cargo these localization patterns include , among others , midcell localization [26 , 36 , 37] , oscillatory movement of ParA and its cargo [24 , 25 , 38] as well as movement from one cell pole to the other [39] . Multiple cargoes are found to equidistantly position along the nucleoid [24–26 , 36 , 37] . To account for the dynamics observed in Par systems , various mechanisms have been proposed . Some models rely on ParA filament formation [24 , 30 , 33 , 40–43] , others challenge this assumption in in vivo systems [39 , 44 , 45] . A diffusion-ratchet mechanism for the movement of ParB-coated beads in vitro and DNA segregation in vivo has been introduced [32 , 45–49] . Based on the observation that DNA has elastic properties [39 , 50] , a DNA-relay mechanism for the movement of the partition complex was proposed [39 , 51] . Here , the force exerted on the cargo is attributed to the elastic properties of the chromosome . Ietswaart et al . observed that if a plasmid is located off-center on the nucleoid , the ParA flux from the left and right sides of the plasmid differ [36] . Based on this idea , they proposed a model that produces equal plasmid spacing over the nucleoid as long as the plasmid moves in the direction of the higher ParA concentration [36] . Additionally , models based on reaction-diffusion equations for Par protein dynamics , have been introduced [52–55] . In order to account for the experimental observations in M . xanthus cells , we have proposed a model for midcell localization [5] that is inspired by , but also differs from previous models for Par systems ( see S1 Text ) . The key experimental observations for the Pom system are: First , PomZ accumulates at the cluster consisting of PomX and PomY proteins , which is in contrast to a low ParA density at plasmids / the partition complex [24 , 47] . Conversely , there are also positioning systems that show an accumulation of the ATPase at the cargo [34 , 56] , which resembles the observations for the Pom system . Second , the cluster is relatively large ( 0 . 7 μm in length , [5] ) compared to plasmids / partition complexes ( about 0 . 1 μm in length , [39] ) . Third , the PomZ proteins diffuse quickly on the nucleoid compared to a slowly moving Pom cluster [5] . Finally , fluorescence micrographs of M . xanthus cells do not show a clear depletion zone in PomZ in the wake of the cluster , in contrast to observations for Par systems [24 , 38 , 47] . The latter observation can be explained by the fast PomZ dynamics on the nucleoid . Our model suggests a positioning mechanism that relies on the biasing of fluxes of PomZ dimers on the nucleoid , similar to the equipositioning mechanism proposed by Ietswaart et al . [36] . With this model we were able to reproduce midnucleoid localization with physiologically relevant parameters [5] , but it remained unclear how the movement of the cluster changes when the rates of the key biological processes involved are varied . Here , we investigate the robustness of Pom cluster dynamics in our model with respect to different parameters , so that our model can be tested by experimentally examining this robustness . Interestingly , we observe that there exists an intermediate ATP hydrolysis rate that minimizes the time the clusters need to reach midnucleoid . Furthermore , we find that fast diffusion of PomZ dimers on the cluster accelerates the movement of the cluster towards midnucleoid . To gain a better understanding of the cluster dynamics observed in the in silico parameter sweeps , we investigate how PomZ dimers generate a net force on the cluster in our model . For the case where the PomZ gradient builds up faster than the velocity of cluster movement , we derive a semi-analytical approximation for the average cluster trajectory , which dissects the generation of a net force into two parts: the difference between the diffusive PomZ fluxes into the cluster from either side , and the force exerted by a single PomZ dimer during its interaction with the cluster . This net force can account for the movement of the cluster to midnucleoid . In contrast , when the PomZ dimers diffuse slowly on the nucleoid , we observe oscillatory cluster movement .
We employ a stochastic lattice gas model , introduced in [5] , to understand the dynamics of the PomXY cluster , i . e . the cluster consisting of PomX and PomY proteins , in M . xanthus bacterial cells [5] . In this model , both the nucleoid and the PomXY cluster are reduced to one-dimensional lattices of length L and Lc , respectively ( Fig 1 ) . Our model consists of two parts: first , the PomZ dynamics , and second the PomXY cluster movement due to its interactions with PomZ . We first describe the PomZ dynamics in the next paragraph . PomZ can occur in different configurations: it can be bound to ADP or ATP and in the latter case , form dimers . In our model we incorporate only the ATP-bound dimeric form of PomZ explicitly . We model the PomZ dimers effectively as springs with spring stiffness k to account for the elastic properties of the chromosome and the PomZ dimers . Each PomZ dimer spring has two binding sites , one which connects to the nucleoid , and one which connects to the PomXY cluster . ATP-bound PomZ dimers can bind with the first binding site to the nucleoid with rate kon ( Fig 1A ( 1 ) ) , except where the PomXY cluster is located , and diffuse on the nucleoid with diffusion coefficient Dnuc ( Fig 1A ( 2 ) ) . Because of thermal fluctuations , the relative position of the second binding site , which enables PomZ to bind to the PomXY cluster , is distributed according to a Boltzmann distribution with the energy of the spring . Therefore , PomZ dimers can attach to the PomXY cluster even if their nucleoid binding sites are not directly below the cluster ( Fig 1A ( 3 ) and 1B ) . We include this factor in the model by multiplying the rate of attachment , k a 0 , by the Boltzmann factor corresponding to the energy of the spring ( as in [57] ) : k a = k a 0 exp [ - 1 2 k k B T ( x i clu - x i nuc ) 2 ] . ( 1 ) The positions of the cluster and nucleoid binding sites of the i-th PomZ dimer bound to the nucleoid and the PomXY cluster are given as x i clu and x i nuc , respectively ( see Fig 1B ) . PomZ dimers bound to the PomXY cluster and the nucleoid are assumed to diffuse on both scaffolds ( Fig 1A ( 4 ) ) . This assumption is motivated by two experimental observations . First , fluorescently tagged PomZ brightly stains the entire cluster in fluorescence micrographs [5] . Second , in a mutant with PomZ dimers that cannot bind to DNA , PomZ is homogeneously distributed inside the cell , which suggests that PomZ dimers only bind to the PomXY cluster when they are nucleoid-bound [5] . Based on these two experimental findings it seems reasonable that PomZ dimers are also mobile on the PomXY cluster as otherwise the concentration of PomZ would be rather concentrated at the cluster edges . PomZ can diffuse on the PomXY cluster and nucleoid with different diffusivities: We assume that the hopping rates are ϵ hop , nuc 0 = D nuc / a 2 and ϵ hop , clu 0 = D clu / a 2 , with the lattice spacing a , respectively , being weighted by a Boltzmann factor that accounts for the energy change of the spring due to the movement: ϵ hop = ϵ hop 0 exp [ - 1 4 k k B T ( ( x i clu , to- x i nuc , to ) 2 - ( x i clu , from- x i nuc , from ) 2 ) ] . ( 2 ) Here , x i clu , from , x i nuc , from and x i clu , to , x i nuc , to signify the position of the binding sites of the i-th PomZ dimer to the cluster and nucleoid before and after hopping , respectively . The additional factor of 1/2 in the exponent is chosen such that detailed balance holds for the hopping events and the rates for hopping to a neighboring site and hopping back are the inverse of each other ( see [57] ) . A PomZ dimer is most likely to move in the direction that relaxes the spring ( cf . exponential factor in Eq 2 ) . We chose reflecting boundary conditions for diffusion of PomZ on both the nucleoid and the PomXY cluster . Next , we discuss the transition from the nucleoid-bound state of PomZ to the cytosolic state . In the experiments , PomX , PomY and DNA stimulate the ATPase activity of PomZ , which leads to a conformational change and finally to detachment of two ADP-bound PomZ monomers from the nucleoid [5] . In our model , we combine the processes of nucleotide hydrolysis and detachment into one rate by assuming that nucleoid- and cluster-bound PomZ dimers are released into the cytosol with hydrolysis rate kh ( Fig 1A ( 5 ) ) . The ADP-bound PomZ monomers must then exchange ADP for ATP and dimerize before they can rebind to the nucleoid . This leads to a delay between detachment from and reattachment to the nucleoid ( Fig 1A ( 6 ) ) . Because of this delay and rapid diffusion of PomZ in the cytosol [5] we assume that upon detachment of PomZ from the nucleoid , it can rebind to any lattice site of the nucleoid with the same probability . The total number of PomZ dimers is assumed to be constant and is denoted by Ntotal . So far we have described the stochastic dynamics of the PomZ dimers . The interactions of PomZ dimers with the PomXY cluster result in forces being exerted on the cluster , which cause it to move . The observable of interest is the cluster position , xc , over time . We approximate the cluster dynamics as overdamped , such that the equation of motion for xc is given by the following force balance equation γ c ∂ t x c = - k ∑ i = 1 N ( x i clu - x i nuc ) , ( 3 ) with γc being the friction coefficient of the PomXY cluster in the cytosol and N the total number of cluster-bound PomZ dimers . Experiments show that the Pom cluster displays very little motion in M . xanthus cells that lack PomZ , whereas its mobility is increased if PomZ is present [5] . Based on this observation , we disregard movements of the cluster due to thermal noise and focus on the stochasticity in the interactions of PomZ dimers with the PomXY cluster , which in turn lead to stochastic forces acting on the cluster . Therefore , we do not include a Langevin noise term in Eq 3 . Our simulations show that the model indeed yields a robust mechanism for stochastic midnucleoid positioning of the PomXY cluster . The underlying mechanism for midnucleoid localization is based on the flux of PomZ on the nucleoid , which can be described as follows . If the PomXY cluster is located to the left of midnucleoid , the average flux of PomZ dimers into the cluster from the right is larger than that from the left [36] ( red arrows in Fig 1A ) . If particles that attach to the cluster typically exert a net force in the direction from which they reached the cluster , the flux imbalance leads to a net force towards the right . For a cluster that overshoots midnucleoid or is already positioned to the right of midnucleoid , the asymmetry in the fluxes is reversed and the cluster moves back towards midnucleoid . Overall , this leads to a self-regulating process that positions the PomXY cluster at midnucleoid . The stochastic simulations show midnucleoid positioning of the PomXY cluster ( Fig 2 , data shown in black ) with physiologically relevant parameters ( S1 Table , for the discussion of the parameters see S1 Text ) . To identify the parameter range that leads to midcell localization and investigate the role of each parameter on the cluster dynamics , we performed broad parameter sweeps . We varied the attachment rate of PomZ dimers to the nucleoid , kon , the binding rate of nucleoid-bound PomZ dimers to the PomXY cluster , k a 0 , the ATP hydrolysis rate of PomZ dimers , kh , and the mobility of PomZ dimers on the nucleoid , Dnuc , and on the PomXY cluster , Dclu , over a broad range ( Fig 2 ) . We never restricted the PomZ dimer density on the nucleoid and bound to the cluster , as we can assume that these densities are low in the wild type situation with a total number of PomZ dimers of Ntotal ≈ 100 [5] . The parameter sweeps show that increasing the attachment rate to the nucleoid , kon , or the binding rate to the PomXY cluster , k a 0 , decreases the time the cluster needs to reach midnucleoid ( Fig 2A and 2B ) . In both cases , the trajectories become independent of the particular parameter when its value exceeds a certain threshold . We conclude that increasing the rate of attachment of PomZ to the nucleoid or the binding of PomZ to the PomXY cluster speeds up the positioning process until an optimum is reached . Next , we consider the effects of varying the rate of ATP hydrolysis by PomZ dimers associated with the PomXY cluster , which is important to maintain the cyclic flux of PomZ dimers between the cytosolic and nucleoid-bound state . This rate also sets the time scale for the interaction of PomZ dimers with the PomXY cluster . The simulations show that decreasing the hydrolysis rate ( kh = 0 . 01 s−1 ) reduces the velocity of the average cluster trajectory towards midnucleoid ( Fig 2C ) . Qualitatively , large hydrolysis rates ( kh = 10 s−1 ) have essentially the same effect ( Fig 2C ) . Hence , there is a hydrolysis rate kh that minimizes the time the cluster needs to reach midnucleoid . Although the average cluster trajectory behaves similarly for large and small hydrolysis rates , we observe that the variance of the cluster distribution over time decreases with increasing hydrolysis rate ( Fig 2C ) . Apart from the ATP hydrolysis rate , we expect the diffusivity of PomZ on the nucleoid to be a crucial parameter for cluster movement , because it determines the time needed for PomZ dimers to explore the nucleoid to the left or right of the cluster . Interestingly , when we reduce the diffusivity of PomZ on the nucleoid in the simulations , the clusters begin to oscillate around the midnucleoid position ( Fig 2D , 2E and S1 Fig ) . Finally , we also decreased the diffusion constant of PomZ dimers on the PomXY cluster , while keeping the diffusion constant on the nucleoid fixed . In this case , the clusters take longer to reach midcell ( Fig 2F ) . In addition to the parameter sweeps shown in Fig 2 , we also considered the PomXY cluster trajectories when the spring stiffness k and the total PomZ dimer number Ntotal are varied . The cluster trajectories do not change significantly when the spring stiffness is altered over one order of magnitude , and an increase in the particle number increases the velocity of cluster movement towards midcell ( S2 Fig ) . Furthermore , we verified that the results do not change qualitatively in an extension of the stochastic model that includes a small detachment rate of nucleoid- , but not cluster-bound PomZ dimers . For very large detachment rates that are above the expected values from experiments midcell localization of the cluster breaks down ( see S1 Text and S3 Fig ) . To summarize , we observed that there exists an ATP hydrolysis rate that minimizes the time taken to reach midnucleoid . The diffusion constant of PomZ on the nucleoid determines whether the PomXY cluster moves towards or oscillates around midnucleoid . Moreover , the clusters move faster towards midcell if PomZ dimers diffuse faster on the PomXY cluster . In the following , we provide first an analytic approach that explains our observations regarding the cluster dynamics when the ATP hydrolysis rate and the diffusion constant of PomZ on the cluster is varied . We then consider the oscillatory cluster dynamics and give an estimate for the onset of oscillations . Our goal is to understand what generates the force behind midcell positioning in our model . We expect that the cyclic flow of PomZ dimers is at the root of this force: PomZ dimers attach to the nucleoid in their active state ( as ATP-bound PomZ dimers ) , diffuse on the nucleoid and are released into the cytosol in their inactive state ( ADP-bound PomZ monomers ) after encountering the PomXY cluster . We describe how the cyclic flow can lead to a net force in the following . We assume that the PomZ dynamics is fast compared to the PomXY cluster dynamics ( adiabatic assumption ) , and approximate the system by a stationary model , i . e . a system with a fixed cluster position . As we neglect exclusion effects on the nucleoid , PomZ dimers can only interact with each other via the PomXY cluster . However , when the cluster is stationary , no interaction between the cluster-bound PomZ dimers is possible , and thus the movements of different PomZ proteins are not correlated . Therefore , we can consider the interactions of PomZ dimers with the PomXY cluster as independent , which yields the following deterministic approximation for the total net force , F , acting on a cluster at position xc F ( x c ) = ( N R ( x c ) - N L ( x c ) ) f , ( 4 ) with f being the time-averaged force exerted by a single PomZ dimer that attaches to the nucleoid on the right side of the cluster . For symmetry reasons , a PomZ dimer coming from the left then exerts a time-averaged force −f . NR and NL denote the numbers of PomZ dimers that are bound to the cluster and had originally attached to the nucleoid to the right and left of the cluster , respectively . These two numbers increase with the diffusive flux of nucleoid-bound PomZ dimers reaching the cluster region from the right and left side , jR/L , respectively , and decrease with the ATP hydrolysis rate , kh , as long as the attachment rate to the PomXY cluster is non-zero . Hence , we obtain N R / L ( x c ) = j R / L ( x c ) k h ( 5 ) in the steady-state . Inserting this into Eq 4 , yields F ( x c ) = j R ( x c ) - j L ( x c ) k h f = f k h j diff ( x c ) ≡ C j diff ( x c ) . ( 6 ) We conclude that the net force is proportional to the flux difference of PomZ dimers at the cluster , jdiff , and the proportionality constant is given by C = f/kh . Importantly , simulation results with a fixed cluster position confirm the observation that the total force exerted on the cluster is proportional to the PomZ flux difference ( S4 Fig ) . Next , we investigate how the net force exerted on the PomXY cluster results in movement of the cluster . Notably , the PomZ dimers interacting with the cluster not only produce a net force on the cluster , they also reduce the mobility of the cluster by tethering it to the nucleoid . We assume that these two processes can be considered independently . We simulated the steady-state velocity with which a cluster moves when a fixed number of PomZ dimers are bound to it and an external force is applied to the cluster ( see Materials and methods section for details ) . We found that this velocity varies linearly with the force ( S5 Fig ) , which suggests that the force exerted on the cluster is balanced by a frictional force with effective friction coefficient γ ( xc ) : F ( xc ) = γ ( xc ) v ( xc ) . With Eq 6 we obtain the central equation of our analysis v ( x c ) = F ( x c ) γ ( x c ) = C j diff ( x c ) γ ( x c ) , ( 7 ) which relates the average velocity of the cluster to the flux difference of PomZ dimers into the cluster , the proportionality constant C and the effective friction coefficient γ ( xc ) of the cluster . To obtain the average cluster trajectory , we need to integrate Eq 7 over time . In the following we derive analytical expressions for the flux difference into the cluster and the effective friction coefficient of the PomXY cluster . The constant C we determine from simulations . Since C does not change with the cluster position , xc , the dependence of the velocity on xc is given by an analytical expression , which can be integrated ( numerically ) . We observe a marked discrepancy between the simulated average cluster trajectory and our approximation when the diffusion constant of PomZ on the nucleoid is reduced and the cluster oscillates around midnucleoid ( Fig 4D and 4E ) . Deviations from our theoretical predictions are to be expected in this situation , because we make an adiabatic assumption in our semi-analytical approach , i . e . we assume that the PomZ dimer dynamics on the nucleoid is fast compared to the cluster movement . This assumption no longer holds when PomZ dimers diffuse slowly on the nucleoid . In this case , the distribution of PomZ density along the nucleoid determined from simulations with a dynamic cluster deviates drastically from its steady-state distribution ( S12 Fig ) . If the cluster initially lies to the left of midnucleoid and approaches midnucleoid from that side , our theory predicts a symmetric PomZ density , whereas the simulations show a higher density in front of the cluster . The flux difference also deviates from the stationary case: it increases as the cluster moves towards midnucleoid instead of vanishing at midnucleoid ( S12 Fig ) . Both the asymmetric density and the non-zero flux difference at midnucleoid are in accordance with the observed oscillatory behavior . The switch between cluster localization at midnucleoid and oscillatory movement around midnucleoid is regulated by the relative time scales of PomZ dynamics and cluster dynamics: If the PomXY cluster is moving slowly or the PomZ dimers move fast , the latter have time to adjust to a change in the cluster position . On the other hand , if the cluster moves fast or the PomZ dimers move slowly , the PomZ dimer distribution deviates from the stationary case . The delay between the movement of the cluster and the build-up of the PomZ gradient , which in turn biases the movement of the cluster , leads to oscillations: the longer the delay , the larger the amplitudes of the oscillations . To investigate the oscillatory case further , we performed additional simulations in which the diffusion constant of the PomZ dimers and that of the PomXY cluster in the cytosol , which is inversely proportional to the friction coefficient , γc , according to the Stokes-Einstein relation , were varied . As expected , we find oscillatory behavior of the clusters for low diffusion constants of PomZ on the nucleoid ( Fig 6 ) . In the oscillatory regime we find both bimodal and monomodal cluster position distributions ( Fig 6 ) . As mentioned above , the onset of oscillations depends on the time scales of PomZ gradient formation and cluster movement . In order to understand how the parameters change the behavior of the cluster trajectory , i . e . lead to oscillatory movement or midcell positioning , we assume that the cluster is located at midnucleoid and search for a stability condition that distinguishes the two behaviors . The diffusion time for a PomZ dimer to explore a nucleoid of size L is given by t PomZ = L 2 D nuc . ( 16 ) In theory , the velocity of a cluster that starts from midnucleoid should be zero , because there should be no difference between the fluxes of PomZ dimers from both sides . However , due to stochastic effects , more particles may attach to the cluster from the right than from the left side , which will displace the cluster to the right . For our time scale argument , we consider an extreme case: we assume that PomZ dimers only arrive from one side , which we choose to be the right side without loss of generality . The time required for a cluster to move the length of the nucleoid is then given by t cluster = L v ≈ L γ ( 0 . 5 L ) C j right ( 0 . 5 L ) , ( 17 ) with jright being the flux of PomZ dimers into the cluster from the right . Here , we approximate the velocity of the cluster by its effective description , Eq 7 , using xc = 0 . 5L , and replace the flux difference with the flux from the right only . According to Eq 17 , the condition for stable positioning of the cluster at midnucleoid t PomZ ≪ t cluster ( 18 ) yields D nuc L ≫ C j right ( 0 . 5 L ) γ ( 0 . 5 L ) = C j right ( 0 . 5 L ) γ c + ( k B T N ( 0 . 5 L ) ) / ( D clu + D nuc ) . ( 19 ) For the parameter sweeps considered before ( Fig 2 and S2 Fig ) , we find tcluster ≫ tPomZ for all cases except for small diffusion constants of PomZ on the nucleoid . With our time-scale argument , Eq 19 , we can make further predictions as to which parameters should result in oscillations . First , we consider a change in the total particle number , Ntotal . Both jright as well as the number of cluster-bound proteins , N , are proportional to Ntotal , and C does not depend on Ntotal . Therefore , the right-hand side of Eq 19 is proportional to Ntotal for small values of Ntotal and converges to a constant for large values . From this we expect that oscillatory behavior may occur for large particle numbers . Simulations with 500 PomZ dimers and a smaller diffusion constant of PomZ on the nucleoid and the PomXY cluster compared to the parameters in S1 Table ( Dnuc = Dclu = 0 . 01 μm2/s ) indeed show oscillatory behavior , whereas simulations with the same parameters , but 100 PomZ dimers show midnucleoid localization ( S13 Fig ) . However , for very large PomZ dimer numbers we expect exclusion effects , which are not considered here , to have an impact that will also affect the cluster dynamics . Second , we investigate the effects on the cluster dynamics when changing the nucleoid length L . Again , the constant C , which represents the force exerted by a single PomZ dimer on the PomXY cluster , does not depend on L . The number of cluster-bound proteins decreases with increasing L , because the relative size of the cluster Lc/L decreases and the total PomZ dimer number in the system is constant . Hence , also the flux of PomZ in the system is reduced , which leads to a decrease of the flux jright with increasing L . Bringing all terms in Eq 19 that depend on L to the right hand side yields a curve that first increases with L , then reaches a maximum and decreases again for large L . Hence , we expect no oscillations for small and large nucleoid lengths and oscillations might occur for intermediate lengths . Simulations with intermediate and large nucleoid lengths L indeed show this behavior ( S13 Fig ) .
We analyzed how the cluster movement changes when the rates for the key biological processes are varied over a broad range . We found that there exists an optimal ATP hydrolysis rate of PomZ such that the time the cluster needs to move to midnucleoid is minimized . A parameter sweep of the diffusion constant of PomZ on the PomXY cluster shows that the mobility of PomZ dimers on the PomXY cluster is important for cluster movement towards midnucleoid . Qualitative changes in the cluster trajectories are observed when the diffusion constant of PomZ on the nucleoid is reduced: midnucleoid positioning of the cluster switches to oscillatory behavior of the cluster around midnucleoid . Hence , we conclude that positioning of the cluster in the flux-based model critically depends on the time scale for the cluster dynamics in comparison to the one for the PomZ dimer dynamics on the nucleoid . If the latter is slow compared to the cluster dynamics , the cluster will oscillate around midnucleoid . In contrast , fast PomZ dynamics on the nucleoid leads to midnucleoid localization of the cluster . In the latter case , we find that the average velocity of the PomXY cluster can be described by the PomZ flux difference into the cluster , which measures how far away the cluster is from midnucleoid , the force exerted by a single PomZ dimer on the cluster , and the effective friction coefficient of the cluster , which depends on the number of PomZ dimers bound to it ( semi-analytical approach ) . This approach allows for further mechanistic insights into the cluster movement by PomZ dimer interactions . With it we can explain the dependence of the cluster dynamics on the model parameters as observed in our simulations . The mechanism we propose for midcell localization of the Pom cluster in M . xanthus is based on a flux-balance argument , which was previously proposed for positioning by the Par system [36] and also for self-organized positioning of protein clusters that dynamically form on the nucleoid [52] . In the model by Ietswaart et al . [36] and the model we present here , the cargo is a fixed structure , whereas Murray and Sourjik [52] consider a reaction-diffusion model for a protein that can form dynamic clusters on the nucleoid , which are positioned by the same protein due to a flux-balance argument . Necessary conditions for flux-based positioning are that the ATPase diffuses on the nucleoid ( faster than the cargo ) and cycles between a nucleoid-bound and cytosolic state [5 , 36 , 52] . Furthermore , the typical length an ATPase diffuses on the nucleoid before it detaches into the cytosol ( without a preceded interaction with the cargo ) has to be sufficiently large compared to the nucleoid length to ensure positioning of a cargo at midcell ( see S3 Fig , S1 Text ) [52 , 58] . How the forces are generated by the ParA-like ATPase to move the cargo ( plasmid , partition complex or protein cluster ) is still under debate . Lim et al . proposed that forces are generated due to the elasticity of the nucleoid [39] , which we also assume here . Alternatively , a chemophoretic force has been suggested . Chemophoretic forces can explain the net movement of catalytic particles in the direction of an increasing concentration of a solute [58] and have also been applied to positioning of cargoes by the Par system [47 , 53–55 , 58] . To what extent a chemophoretic force and / or the elasticity of the nucleoid lead to the net force that moves the cargoes remains to be investigated . One important experimental observation that differs between the Pom system and several Par systems is that PomZ dimers accumulate at the cluster . In our model , we make two important model assumptions that affect the density profile of PomZ at the cluster: First , we assume that cluster-bound PomZ dimers can only detach from the cluster via ATP hydrolysis , such that the dimers are captured at the cluster until they are released into the cytosol . Second , we assume that cluster-bound PomZ dimers can diffuse on both the cluster and the nucleoid . These assumptions have important implications on how forces are generated at the cluster in our model . We find that the PomZ dimer springs not only exert forces when they attach to the cluster in a stretched configuration ( as in the DNA-relay model , [39] ) , but instead forces can be generated every time a cluster-bound PomZ dimer encounters the cluster’s edge . Our simulations show that the latter contribution to the overall force of a single PomZ dimer is much more important than binding in a stretched configuration , for the parameters we consider . This is in stark contrast to the situation in the DNA-relay model , where only the initial deflection of the ParA dimer from its equilibrium position when binding to the cargo accounts for the generated force . Another important difference between our model and previously proposed models for the Par system that include the elasticity of the nucleoid [39 , 48 , 51 , 59] is how mobile the ATPase is compared to the cargo . In contrast to the aforementioned Par models , the ATPase ( PomZ ) diffuses rapidly on the nucleoid and the cargo ( Pom cluster ) only moves due to its interactions with PomZ dimers , in our model . Fast diffusion of PomZ dimers on the nucleoid and the relatively large spring stiffness of PomZ dimer springs explain the small force exerted on the cluster due to the initial deflection of the spring: this deflection is only small and quickly reduced by diffusion of PomZ . We conclude that force generation based on the elasticity of the nucleoid can be sufficient for cargo translocation even if the mobility of the transporting proteins is higher than the mobility of the cargo . Our observation of an oscillatory cluster movement when the dynamics of the PomZ dimers is slow compared to the dynamics of the PomXY cluster is in agreement with findings for the Par system [48 , 53] , despite differences between their models and ours . Similar to our finding that an intermediate ATP hydrolysis rate of PomZ minimizes the time the cluster needs to reach midcell , Hu et al . observed that an intermediate detachment rate of the ATPase from the cargo leads to the most persistent movement of the cargo [59] . However , their model differs from our model as they consider the movement of a fast diffusing cargo on a two-dimensional DNA-carpet to mimic an in vitro Par system [47] . In contrast , our model for the in vivo Pom system accounts for the nucleoid as an object of finite size . Since the Pom cluster diffuses slowly compared to the PomZ dimers , the diffusive fluxes of PomZ into the cluster need to be accounted for when determining the dependence of the cluster dynamics on the ATP hydrolysis rate ( Fig 5 ) . The model we present here yields a mechanistic understanding of midcell localization of the Pom cluster . So far , not all model parameters are determined experimentally in M . xanthus cells . Hence , it would be important to measure the remaining biological rates , such as the nucleoid attachment rate , the diffusion constants and the cluster binding rate in vivo . Another limitation of our current model is that it is one-dimensional . How the cluster dynamics changes in a three-dimensional geometry is an interesting question for further research . Furthermore , in the current model we do not account for the PomXY cluster formation , but consider the cluster as a fixed structure . This is motivated by the experimental finding that PomX forms filaments in vitro and a high fraction of fluorescently labelled PomX was observed in the cluster in vivo [5] . However , it remains unclear how the cluster is formed in vivo and how the size of the cluster is maintained from one cell generation to the next . Our model for the Pom cluster positioning makes three important predictions , which would be interesting to test experimentally: First , the cluster starts to oscillate if PomZ dimers diffuse slowly on the nucleoid . We hypothesize that this might be tested experimentally by increasing the binding affinity of PomZ dimers to the DNA and in this way decreasing the mobility of PomZ on the nucleoid . Second , we predict that there is an optimal ATP hydrolysis rate to minimize the time the cluster takes to reach midnucleoid . Decreasing the rate of ATP hydrolysis by PomZ dimers associated with the PomXY cluster in experiments reduced the velocity of cluster movement towards midcell [5] . It would be interesting to test whether the velocity of the cluster is also reduced for an enhanced ATP hydrolysis rate in in vivo experiments . Finally , we predict that the mobility of the PomZ dimers on the Pom cluster can increase the velocity of the cluster movement . To test this model prediction , experiments to uncover the dynamics of PomZ dimers bound to the cluster are needed . The research presented here gives insights into the dynamics of the Pom cluster in M . xanthus , which is determined by its interactions with the nucleoid-bound PomZ dimers . With our semi-analytical approach we gain a better mechanistic understanding of the net force generation in our model . This approach might also prove to be useful for the related ParABS systems or other stochastic , out of equilibrium systems to position intracellular cargoes .
In the simulations to determine the cluster dynamics , all PomZ dimers are initially in the cytosol . The PomXY cluster position is kept fixed for tmin = 10 min with all possible actions of the PomZ dimers allowed . As a result , the initial condition resembles the stationary distribution of PomZ dimers . The initial position of the cluster is such that the left edge of the cluster and the nucleoid coincide . To derive PomZ flux and density profiles at specific cluster positions , the simulated fluxes and densities are recorded only if the cluster is within a certain distance of a predefined position of interest . For example , to get the PomZ flux / density for clusters at 20% nucleoid length , recording begins when the PomXY cluster is in the region 20 ± 0 . 2% and stops if it leaves the region 20 ± 1% . Averaging is performed over all times at which the cluster resides within the specific region , weighting each density or flux profile with the corresponding time spent by the cluster in that specific region . To estimate the difference in PomZ flux into the PomXY cluster from either side , the maximal and minimal flux values in the average flux profile of PomZ dimers bound to the nucleoid , but not the PomXY cluster , are determined . These values are typically found a short distance from the edge of the PomXY cluster region , because PomZ dimers can attach to the cluster in a stretched configuration . The two extreme flux values of opposite sign are added together to get the average flux difference of PomZ dimers into the cluster . Simulations with a fixed position of the PomXY cluster are performed to measure the force exerted by a single PomZ dimer on the cluster ( “one-particle simulations” ) or to measure the PomZ dimer flux into the cluster and the forces exerted on the cluster for an arbitrary number of PomZ dimers in the system . In these simulations , the PomZ dimer ( s ) are initially in the cytosol . When the adiabatic assumption holds true , the results from the stationary cluster simulations can be used as approximations for the PomZ dynamics in the dynamic cluster simulations . | In order for the rod-shaped bacterium M . xanthus to reproduce , its genetic content must be duplicated , distributed equally to the two cell halves and then the cell must divide precisely at midcell . Three proteins , called PomX , PomY and PomZ , are important for the localization of the cell division site at midcell . PomX and PomY form a cluster and PomZ tethers this cluster to the bacterial DNA or nucleoid ( region containing the chromosomal DNA ) and is important for the movement of the cluster from the nucleoid pole towards midcell . We are interested in the question how the cluster trajectories change when the PomZ dynamics is varied . To address this question we investigate a previously developed mathematical model that incorporates the nucleoid , the cluster and PomZ . We simulated the cluster trajectories for different model parameters , such as different diffusion constants of PomZ on the nucleoid . Interestingly , when PomZ diffuses slowly on the nucleoid , we observed oscillatory cluster movements around midcell . Our results provide general insights into intracellular positioning of proteins . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"dimers",
"(chemical",
"physics)",
"classical",
"mechanics",
"enzymes",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"enzymology",
"phosphatases",
"simulation",
"and",
"modeling",
"research",
"and",
"analysis",
"methods",
"proteins",
"chemistry",
"adenosine",
"triphosphatase",
"physics",
"biochemistry",
"hydrolysis",
"cell",
"biology",
"biology",
"and",
"life",
"sciences",
"chemical",
"reactions",
"physical",
"sciences",
"chemical",
"physics",
"cytosol",
"atp",
"hydrolysis"
] | 2018 | Regulation of Pom cluster dynamics in Myxococcus xanthus |
All State TB control programmes in Nigeria were requested to submit 25–50 smear-positive Ziehl-Neelsen ( ZN ) stained slides for screening during 2013–2014 . DNA was extracted from 929 slides for spoligotyping and drug-resistance analysis using microbead-based flow-cytometry suspension arrays . Spoligotyping results were obtained for 549 ( 59 . 1% ) of 929 samples . Lineage 4 Cameroon sublineage ( L4 . 6 . 2 ) represented half of the patterns , Mycobacterium africanum ( L5 and L6 ) represented one fifth of the patterns , and all other lineages , including other L4 sublineages , represented one third of the patterns . Sublineage L4 . 6 . 2 was mostly identified in the north of the country whereas L5 was mostly observed in the south and L6 was scattered . The spatial distribution of genotypes had genetic geographic gradients . We did not obtain results enabling the detection of drug-resistance mutations . We present the first national snapshot of the M . tuberculosis spoligotypes circulating in Nigeria based on ZN slides . Spoligotyping data can be obtained in a rapid and high-throughput manner with DNA extracted from ZN-stained slides , which may potentially improve our understanding of the genetic epidemiology of TB .
In 2015 , six countries , Nigeria , India , Indonesia , China , Pakistan and South-Africa , accounted for 60% of all TB cases in the world , with Nigeria having the second largest TB burden in Africa , with an estimated 586 , 000 cases in 2015 [1] . Nigeria is also among the ten countries with the largest gaps between notifications of new and relapse ( incident ) TB cases and the best estimates of TB incidence in 2016 [2] . Despite the large number of cases , there is a paucity of information of the genetic diversity of Mycobacterium tuberculosis complex ( MTC ) circulating in the country and in west Africa [3–7] . In a previous study , we reported the genetic diversity of MTC in three Nigerian cities ( Abuja , Ibadan and Nnewi ) using a CRISPR-based fingerprinting method ( spoligotyping ) based on DNA extracted from clinical isolates [5 , 8 , 9] . Lineages 4 . 6 . 2 ( Cameroon family ) and Lineages 5 and 6 ( Mycobacterium africanum west African 1 and 2 ) were the main lineages found [5] . The study however reported significant variations between the cities and it was not representative of whole Nigeria . Although it would be desirable to conduct genetic diversity studies on MTC cultures , with a wider geographical representation of patients , this would require considerable resources and would be logistically complex . Hence , in a large country like Nigeria , obtaining a representative set of cultures from all inhabited regions is difficult to achieve . It has been reported that it is possible to conduct spoligotyping analysis directly from Ziehl-Neelsen ( ZN ) stained slides , scratching the material from the slides and extracting the DNA [10] . However the method has a relatively poor sensitivity and has not gained wide acceptance as a direct molecular identification technique [10] . Despite this limitation , direct spoligotyping from smears would be logistically simpler and facilitate studies with wider geographical representation , bypassing the need for culture , and allowing the direct identification of M . africanum ( L5 and L6 ) , which has specific metabolic requirements and is difficult to grow [11 , 12] . In one recent study we demonstrated that the apparent disappearance of M . africanum from Burkina Faso was a sampling artefact [13 , 14] and other teams in Cameroon and Benin are currently facing the dilemma on whether M . africanum is truly disappearing or not , and a study in Ghana has shown that M . africanum still represents 20% of all TB cases [15] . Indeed , an alternative explanation would be the progressive replacement of M . africanum by more modern ( Lineage 4 ) isolates [14] . The aim of this study was to conduct a nation-wide description of the diversity of tuberculosis genotypes based on sputum smear extracts , to assess the global geographical genetic structure of MTC in Nigeria . In addition , we tested a new Nucleic Acid Amplification test ( NAAT ) method «TB-SPRINT» , that allows simultaneous spoligotyping and prediction of drug-resistance in a sample collection based on ZN extracted DNA [16–18] .
Since we wanted to provide a detailed spatial analysis of circulating MTC genotypes , we studied M . tuberculosis diversity in 36 states using ZN-stained smears representative of the country to build geographical genetic maps . A total of 929 ZN slides with grades 3+ ( n = 505 ) and 2+ ( n = 424 ) , were collected from adults attending health facilities with cough of more than two weeks duration who had not received treatment for TB [19] . Samples were gathered as part of routine clinical work for the diagnosis of TB . Slides processed by the laboratories were then selected for the routine external quality assurance ( EQA ) activities conducted by the National TB and Leprosy Control Program ( NTLCP ) to evaluate the quality of smear microscopy in each State . These slides are routinely anonymised at the time they are submitted to the reference laboratory . Smear positive samples received by the reference laboratory for blind rechecking as part of the EQA were then selected for the study . Only date and origin of the sample were collected . Subjects did not provide informed consent as this is a routine diagnostic procedure . The Research Ethics Committee of the Liverpool School of Tropical Medicine and the Institutional Review Board of Zankli Medical Centre approved the study protocol . ZN slides were requested via the State TB focal person of the NTLCP from all 36 States and the Federal Capital Territory of Nigeria during the years 2013 and 2014 . Slides were shipped by post to Zankli Medical Centre in Abuja . ZN staining and smear grading were performed by the routine diagnostic laboratories following national guidelines [20] . DNA extraction was performed from the stained slides as follows: the mineral oil was removed with xylene , 25 μl of filtered TRIS-EDTA was added to the slides and the material was scraped off into a microcentrifuge tube . The tubes were shipped at room temperature to the Hospital Universitari Germans Trias i Pujol ( Spain ) , where 75 μl of Chelex suspension was added . After thorough mixing , samples were incubated for 30 minutes at 95°C , sonicated for 5 minutes and centrifuged for 15 minutes at 14 , 000 g at 4°C . The supernatant was transferred to fresh microcentrifuge tubes and sent to the Institute for Integrative Cell Biology ( Gif-sur-Yvette , France ) for high-throughput spoligotyping . High-throughput spoligotyping was performed using the microbead-based method using flow-cytometry suspension arrays ( Luminex 200 , Luminex Corp , Austin , TX ) , as previously described [21] . TB-SPOL and TB-SPRINT kits were purchased from Beamedex ( Beamedex , Orsay , France; www . beamedex . com ) . Given the low DNA content of the samples , the PCR protocol was adjusted by increasing PCR cycles from 20 to up to 35 cycles . Each spoligotyping profile was verified at least twice and doubtful profiles were rejected . Quality control was conducted by three experts ( MKG , BMM , CS ) who independently assessed individual spoligotyping profiles . Full or empty squares represents the presence or absence of the 43 classical CRISPR spacers . The spoligotyping patterns were further labelled using the SITVITWEB database [22] . Lineages ( L1 to L7 ) and sub-lineages ( such as L . 4 . 6 . 2 , the Cameroon Lineage ) were designated using the latest taxonomical standards as provided by whole-genome sequencing , using published lineage definition [23 , 24] . When available , the lineage designation was complemented by the classical spoligotyping-based clade designation , e . g . Lineage 4 . 6 . 2 is also designated as the «Cameroon Clade ( CAM ) » [25] . An Excel result file was imported into Bionumerics version 7 . 5 ( Biomérieux , Applied Maths , St Martens-Latem , Belgium ) . A minimum spanning tree ( MST ) was built using the Bionumerics user’s manual ( Fig 1 ) . A Mac OSX Version of QGIS ( v2 . 18 Las Palmas de Gran Canaria ) was downloaded and installed from: http://www . kyngchaos . com/software/qgis . A licence-free Nigerian map with administrative level 1 was downloaded as a shapefile from http://maplibrary . org/library/stacks/Africa/Nigeria/index . htm; Latitude and longitude of the main Nigerian cities was downloaded from http://www . downloadexcelfiles . com/wo_en/download-list-latitudelongitude-cities-nigeria# . WUPbwYXSWDQ and http://simplemaps . com/data/world-cities . Nigerian Population records and density was obtained by collecting data from: http://www . iplussolutions . org/isolutions-leads-consortium-streamline-patient-access-essential-treatments-nigeria-0 ( assessed on November 2017 , 20th ) and https://www . citypopulation . de ( assessed on December 2017 , 8th ) . Maps ( Fig 2 and Fig 3 ) were produced using QGIS after importation of . shp and . csv files using the user’s manual ( cf . S2 Table ) .
M . tuberculosis spoligotyping results were obtained in 549 of 929 smear samples ( genotyping rate = 59 . 1% ) . Of these , 327 ( 64 . 8% ) were obtained of the 505 slides with smear grades 3+ and 222 ( 52 . 4% ) of the 424 slides with smear grades 2+ . Spoligotyping patterns were named using Spoligotyping International Type ( SIT ) tags ( S1 Table ) . An MST tree that describes the global population structure is shown in Fig 1 . One hundred and two different spoligotyping patterns were observed , among which 55 had unique patterns with one sample and 47 were clusters with two or more samples . The global distribution of patterns is dominated by the «Cameroon» family , Lineage 4 . 6 . 2 , including SIT61 and derived signatures characterized by the absence of spacers 23–25 and 33–36 , representing about 50% of patterns , and M . africanum , which included M . africanum west African 1 ( L5; SIT431 and SIT338 , signature: absence of spacer 8–12 and 37–39 ) and M . africanum west-African 2 ( L6; SIT181 , signature: absence of 8–9 and 39 ) , which represented approximately 20% of patterns . All other lineages L1 to L4 , including East African India/EAI ( L1 , absence of spacers 29–32 and 34 ) , Beijing ( L2 , absence of spacers 1–34 ) , Central Asia/CAS ( L3 , absence of spacers 4–7 and 23–34 ) , Euro-American ( L4 , absence of spacer 33–36 ) , i . e . all «T spoligotypes» and other derived sublineages ( Haarlem , LAM , S , and others , absence of spacers 33–36 plus specific signatures ) , represented circa 30% of all patterns . The main spoligotyping results are shown in S1 Table with Lineage assignation . Fig 1 shows the cluster analysis . Two hundred and eighty six ( 52 . 1% ) of the 549 samples belonged to the L4 . 6 . 2 ( Cameroon Family ) , of which 278 were found in 13 different clusters ( SIT61 , n = 232; SIT838 , n = 6; SIT844 , n = 3; SIT852 , n = 8; SIT1204 , n = 3; SIT2550 , n = 3; NEW “Koro-Koro 2013” , n = 6 ( described in Cameroon as CAM57 ) ; «NEW1» , n = 4; «NEW12» , n = 2; «NEW13*» , n = 5; «NEW14*» , n = 2; «NEW2» , n = 2; «new-f*» , n = 2 ) , and eight patterns were orphan with a classical L4 . 6 . 2 . signature . Full results are shown in S1 Table and S2 Table . The L5 ( M . africanum west African 1 ) genetic diversity included 87 ( 15 . 8% ) isolates , of which 75 were identified in 11 clusters ( SIT319 , n = 6; SIT320 , n = 9; SIT331 , n = 23; SIT438 , n = 11; SIT856 , n = 7; «NEW11*» , n = 5; «NEW4» , n = 2; «NEW6» , n = 2; «NEW7’*» , n = 2; «NEW8*» , n = 6; « NEW9 » , n = 2 ) , and 12 were orphan . The L6 ( M . africanum west-African 2 ) included 15 ( 2 . 7% ) samples of which 10 were found in two clusters ( SIT181 , n = 8; «NEW3» , n = 2 ) and five were orphan . The L4 ( T1 subfamilies , i . e . other than L4 . 6 . 2 and L4 . 6 ) included 82 samples among which 75 were found in four clusters ( SIT53 , n = 68; SIT291 , n = 3; SIT334 , n = 2; SIT1580 , n = 2 ) and seven were orphan . Within the T2 subfamily ( Lineage 4 . 6 ) , 12 samples were identified in three clusters ( SIT52 , n = 5; SIT742 , n = 2; SIT848 , n = 3 ) and two patterns were orphan . Lastly , a single pattern of the T3 subfamily and four additional orphan patterns were classified as belonging to the broad L4 lineage . Thirteen samples were classified as H3 ( L4 . 1 . 2 or L4 . 5 ) [24] . Nine of these samples were found in three clusters ( SIT49 , n = 5; SIT50 , n = 2; SIT307 , n = 2 ) and four patterns were orphan . A total of seven samples were classified as belonging to Lineage 4 . 3 ( LAM ) , of which five were in two LAM9 clusters ( SIT42 , n = 3; SIT1064 , n = 2 ) , one was orphan , and one was classified as LAM3 . Twenty samples were classified as Unknown ( «U» ) , among which 16 were found in three clusters ( SIT1869 , n = 2; «NEW10» , n = 2; «NEW5» , n = 14 ) and two were orphan patterns . Finally , the genetic diversity of the remaining 19 samples was as follows: L1 ( EAI5; n = 2 ) , L1 ( EAI2_Manilla; n = 1 ) ; L2 ( Beijing; n = 1 ) , L3 ( CAS1_Delhi; n = 1 ) , L4 . 1 . 1 . 2 ( X , SIT119; n = 3 ) , Lineage 4 ( Manu_ancestor SIT523; n = 3 ) L6 ( Mycobacterium bovis; n = 2 ) , and five orphan patterns with no SIT or family that could be assigned by spoligotyping . Fig 1 presents a Minimum Spanning Tree of all spoligotyping results . Relative diameters of clusters represent the relative percentage of each spoligotyping cluster . Clonal complexes , i . e . samples historically linked to a progenitor clone are easily identified and we observed that L . 4 . 6 . 2 predominates . The other main clusters presented , belonging to L4 and L5 are also mentioned . The relative position of some clusters of L5 on top of Fig 1 is misleading since these clusters are not related to L4 . The States information for the clustered isolates ( Fig 1 , right panel ) shows both intra-State and inter-State clusters . Fig 2A presents previous results in settings where investigations had been done . Fig 2B shows the list of States of Nigeria . Fig 2C and 2D described population density and absolute population data respectively ( see also S2 Table ) . Fig 3A shows the relative percentage of L1-L6 by State . L4 . 6 . 2 was found in 36 out of 37 States ( all except Anambra State ) . L5 was found in 28 States ( 23 in the south and 5 in the north of the country ) . L6 was found in 10 States throughout the country but underrepresented in the south , where L5 predominates . Fig 3A and 3B show that there is a predominance of L4 . 6 . 2 in the north and a predominance of L5 in the south of the country . In six states , all in the north , central , or west ( Yobe , Gombe , Bauchi , Katsina , Kaduna , Niger and Sokoto ) we did not detect any M . africanum . The spatial distribution of L4 . 6 . 2 in Fig 3B shows that this lineage is prevalent everywhere in the country . However , within L4 , L4 . 6 . 2 was relatively less frequent compared to other L4 sublineages where L5 predominates . L5 was present in 28 out of 37 states and was most prevalent in the south and south-east , with Enugu ( 75% , 15/20 ) Imo ( 66% , 4/6 ) Rivers ( 60% , 13/22 ) , Benue ( 47% , 8/17 ) , Akwa Ibom ( 42% , 3/7 ) , Delta ( 35% , 7/20 ) , Ebonyi ( 27% , 6/22 ) and Abia States ( 21% , 4/19 ) , in decreasing order , and a marked south-east tropism ( Fig 3A ) , with Ekiti ( 33% , 5/15 ) and Ondo ( 15% , 2/13 ) being the two most south-western states where L5 clusters were detected . Rivers was the only state with three distinct L5 clusters ( SIT430 , SIT331 and NEW11 ) . Benue and Enugu States show the presence of relatively large L5 clusters ( SIT856 ( n = 6 ) and NEW5 ( n = 13 ) , respectively ) . L6 was found scattered from north to south and east to west in 10 states , detected at low prevalence and only in the Ebonyi state in the south-east . In three states in the north ( Kebbi , Zamfara , Kano ) and four in the south ( Osun , Edo , Abonyi , Taraba ) both L5 and L6 cases were detected .
Nigeria is one of the most ethnically diverse and demographically and economically active countries in Africa with an estimated population of 173 million in 2013 and a life expectancy of 55 years ( 2013 ) [26] . It is also a Federal country with a rich historical , cultural , ethnical , economical and anthropological diversity and many local languages . Nigerian M . tuberculosis genetic diversity had never been investigated at a national scale and recent publications attempting to describe the molecular epidemiology of TB included only a selection of States ( e . g . Anambra , Cross River , Oyo , Plateau , and the FCT ) [4 , 5 , 27–29] . Our study confirms that L4 . 6 . 2 is spread over the whole country . Further whole genome sequencing ( WGS ) would also allow to estimate divergence times and phylogeny within and between sublineages by estimating molecular clock rates based on single nucleotide polymorphisms ( SNPs ) numbers detected [30 , 31] . These new results confirm our previous quantitative results obtained on MTC Lineages prevalence in Anambra , FCT and Oyo states and extends it to the whole Nigerian territory [32] . It is also evident that the L4 . 6 . 2 sublineage is not limited to Cameroon and has a much wider geographic distribution [25 , 33] . Our results suggest that L6 . 4 . 2 is spreading and hence should now be the focus of WGS studies in Nigeria , whether for molecular epidemiology or to gain further knowledge on the resistome and evolutionary history of this L4 sublineage . We found a relatively high percentage of L5-L6 , especially in the south-east of the country , compared to neighbouring countries such as Cameroon and Benin , where it has been suggested that L5-L6 were disappearing [12 , 34] . L5-L6 are geographically restricted and clustered in some states , especially in the south-east , where it represents an important percentage of cases . When comparing to neighbouring Benin and Cameroon on the west and east Nigerian borders respectively , the TB genetic structure is quite different . In Benin , a recent study of 100 isolates recruited in 2014 , reported the following distribution: L1: 0% , L2: 8% , L3: 1% , L4: 77% ( 46% of L . 4 . 6 . 2-CAM and 31% of others ) , L5: 12% , L6: 2% [34] . If we take into account the most recent results obtained , there might be an underestimation of L5 in the Benin study due to difficulty to grow M . africanum [35] . In Cameroon , the prevalence of M . africanum would have dropped in 2004–2005 to 3 . 3% [12] . Another more recent study performed in 2009 on 509 patients in the Adamaoua region reports 2 , 3% of M . africanum [36] . L5 clones ( SIT438 , SIT331 , SIT319 ) are known to be more prevalent in countries around the Gulf of Guinea [6 , 34 , 37] . M . africanum prevalence remains stable in Ghana and it has been found significantly more common in patients of the Ewe ethnic group in this country [38] . Furthermore , there seems to be a high diversity of M . africanum in Ghana [39] , which was not the result of a single outbreak , as the spoligotyping patterns from Ewe patients were quite diverse [38] . In this paper , we do not provide a formal analysis such as spatial regression proving the link of L5 to specific human subpopulations . Nevertheless , we suggest linguistic , ethnical and cultural association of some MTC genotypes ( Lineage 5 ) with specific populations in Nigeria , that shows , as it had been shown in Ghana , a long-lasting co-evolution of M . tuberculosis and Homo sapiens sapiens [15 , 40] . Recently , a study described 315 TB cases caused by M . africanum ( L5-L6 ) in the USA during 2004–2013 [41] . Half of the L5 cases were linked to Nigerian-born patients [41] . We first tagged these spoligotypes using SITVITWEB [22] . Out of 305 patterns , 172 got a SIT label . Common clusters as well as seven new clusters were found to be shared between the USA 2004–2013 study and this present study ( S3 Table ) . Quite often , cases depicted as being unique ( orphan ) in the USA matched with clustered cases found in this study ( S3 Table ) . These results provide clues for further epidemiological and transmission studies in relation to immigration and TB history in the USA . They also demonstrate the quality of spoligotypes obtained on sputum extracts . The limitations of this study include that culture is more sensitive than smear microscopy . However , this is not feasible presently in most of Nigeria . Transporting sputa for cultures reduces this sensitivity due to death of mycobacteria and overgrowth of contaminating flora . We also included a low number of samples collected per State ( median of 25 , range 13–32 ) , which together with the suboptimal performance of spoligotyping from ZN slides and the low number of samples with results per State ( median 16 , range 3–23 ) reduced the power of the study and the possibility to conduct a more detailed geographical analysis at sub-State level . Concerning the low sensitivity of the spoligotyping on ZN extracted material , 60–65% , the classical hot ZN is said to result in lower DNA quality than the modified cold Kinyoun method . However we did not compare the quality and quantity of DNA recovered , as the former method ( hot ZN ) is used throughout the study . Another limitation was the lack of results on drug-resistance-linked SNPs ( rpoB; katG , inhA ) after PCR amplification by TB-SPRINT on the same material , a limitation that was again likely due to a suboptimal DNA extraction method . Partial results only were obtained with poor sensitivity and paucibacillary DNA-containing material , is likely to require improved DNA extraction methods such as the selective target enrichment using specific oligonucleotide coupled microspheres or other more sophisticated DNA extraction procedures [42–43] . Despite these limitations , this study generated the first nation-wide genetic diversity study of MTC in Nigeria . In three states , Oyo , FCT , Anambra this study confirms our previous quantitative results on the L4/L5-L6 relative prevalence [5] . Our results on L5-L6 prevalence are informative of the phylogeography of M . africanum in Nigeria . In addition , our high-throughput spoligotyping technique has the potential to generate more country-wide data of the genetic distribution of MTC in countries with limited resources . This approach improves the detection of M . africanum , which is often under-represented in culture-based studies [35] . With the advent of Next Generation Sequencing ( NGS ) , more precise population-based information could easily be obtained at affordable costs , without TB culture , even in Africa . Such projects may however require samples either with a higher DNA yield or more complex DNA extraction procedures , and necessarily result in a more selected sample of patients attending secondary and tertiary hospitals . The use of slides in turn generates genetic epidemiological information at a larger scale without the need of this infrastructure . Last but not least , our results suggest that some specific TB control and/or preventive measures could be specifically taken in densely populated areas in the south-east region of the country ( Fig 2C and 2D ) , and that assessing the drug-resistance status of the L4 . 6 . 2 lineage is an important issue for the Nigerian TB control program . | Using a classical genotyping method designated as “spoligotyping” , which targets polymorphic repetitive DNA loci , we present a general snapshot of the genetic diversity of Mycobacterium tuberculosis complex causing tuberculosis in Nigeria . Our results were obtained on a collection of 549 DNAs , extracted from Ziehl-Neelsen stained sputum smears gathered and representative from 36 Nigerian states and during 2013–2014 . We show the ubiquitous presence on the Nigerian territory of a sublineage of Lineage 4 , designated as L4 . 6 . 2 or “Cameroon” Lineage , which represents almost 50% of all patterns , more prevalent where Mycobacterium africanum west African 1 ( Lineage 5 ) is absent . We also show that Lineage 5 is geographically linked to the south-east of the country , where it is the most prevalent , and represents approximately 20% of all the samples . The last third is linked to all other L4 sublineages and to L6 ( M . africanum west African 2 ) . Our results confirm the strong phylogeographical structure of Mycobacterium tuberculosis complex in Nigeria and are suggestive of a long-term coevolution history between Homo sapiens sapiens and Mycobacterium tuberculosis complex . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"biogeography",
"ecology",
"and",
"environmental",
"sciences",
"medicine",
"and",
"health",
"sciences",
"population",
"dynamics",
"population",
"genetics",
"geographical",
"locations",
"tropical",
"diseases",
"ethnicities",
"bacterial",
"diseases",
"population",
"biology",
"bacteria",
"extraction",
"techniques",
"africa",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"geography",
"cameroon",
"tuberculosis",
"actinobacteria",
"phylogeography",
"nigeria",
"people",
"and",
"places",
"dna",
"extraction",
"african",
"people",
"mycobacterium",
"tuberculosis",
"earth",
"sciences",
"genetics",
"biology",
"and",
"life",
"sciences",
"population",
"groupings",
"evolutionary",
"biology",
"organisms",
"geographic",
"distribution"
] | 2018 | Mycobacterium tuberculosis complex genotypes circulating in Nigeria based on spoligotyping obtained from Ziehl-Neelsen stained slides extracted DNA |
We combined the Hodgkin–Huxley equations and a 36-state model of gap junction channel gating to simulate electrical signal transfer through electrical synapses . Differently from most previous studies , our model can account for dynamic modulation of junctional conductance during the spread of electrical signal between coupled neurons . The model of electrical synapse is based on electrical properties of the gap junction channel encompassing two fast and two slow gates triggered by the transjunctional voltage . We quantified the influence of a difference in input resistances of electrically coupled neurons and instantaneous conductance–voltage rectification of gap junctions on an asymmetry of cell-to-cell signaling . We demonstrated that such asymmetry strongly depends on junctional conductance and can lead to the unidirectional transfer of action potentials . The simulation results also revealed that voltage spikes , which develop between neighboring cells during the spread of action potentials , can induce a rapid decay of junctional conductance , thus demonstrating spiking activity-dependent short-term plasticity of electrical synapses . This conclusion was supported by experimental data obtained in HeLa cells transfected with connexin45 , which is among connexin isoforms expressed in neurons . Moreover , the model allowed us to replicate the kinetics of junctional conductance under different levels of intracellular concentration of free magnesium ( [Mg2+]i ) , which was experimentally recorded in cells expressing connexin36 , a major neuronal connexin . We demonstrated that such [Mg2+]i-dependent long-term plasticity of the electrical synapse can be adequately reproduced through the changes of slow gate parameters of the 36-state model . This suggests that some types of chemical modulation of gap junctions can be executed through the underlying mechanisms of voltage gating . Overall , the developed model accounts for direction-dependent asymmetry , as well as for short- and long-term plasticity of electrical synapses . Our modeling results demonstrate that such complex behavior of the electrical synapse is important in shaping the response of coupled neurons .
In most models of neuronal networks , it is assumed that electrical synapses exhibit constant conductance , and that electric synaptic transmission is bidirectional and symmetric . However , experimental studies show that these assumptions are not always satisfied . For instance , some synapses formed of gap junction channels exhibit an instantaneous conductance–voltage rectification , which promotes a direction-dependent asymmetry of electrical signaling [1–4] . In addition , all members of the connexin ( Cx ) family forming gap junction channels exhibit a sensitivity of junctional conductance to the transjunctional voltage [5] . Moreover , voltage sensitivity of gap junctions can be strongly affected by chemical factors , e . g . by intracellular concentrations of H+ , Ca2+ or Mg2+ [6–8] . Thus , electrical synapses are not just passive pores , but can exhibit dynamic changes of junctional conductance . Presumably , these changes in electrical synaptic strength could affect the transfer of an electrical signal . The purpose of our study was to develop a computational model for evaluation of such an interaction between electrical synapses and signal transmission between coupled neurons . The first quantitative models describing equilibrium [9 , 10] and kinetic [11] properties of junctional conductance dependence on transjunctional voltage were based on the assumption that the channel can be in two states , open and closed . Later , single channel studies have shown that transjunctional voltage causes channels to close to a subconductance ( residual ) state [12 , 13] with fast gating transitions , and to a fully closed state with slow gating transitions [14 , 15] . Thereafter , it was proposed that gap junction channels comprise two types of gating mechanisms , fast and slow , each exhibiting rectification of their unitary conductances depending on the voltage across them . These properties were described in a stochastic 16-state model ( 16SM ) of gap junction channel gating [16] in which fast and slow gates operate between open ( o ) and closed ( c ) states . However , experimental data from our and other groups [17 , 18] allowed us to suggest that the slow gate operates between open ( o ) , initial-closed ( c1 ) and deep-closed ( c2 ) states . Such a suggestion was implemented in a 36-state model ( 36SM ) of voltage gating [19] . The 36SM allowed us to reproduce experimentally observed gating behavior of gap junction channels more adequately than 16SM , especially regarding the kinetics of conductance recovery , or a low fraction of functional channels clustered in junctional plaques . Earlier [20] , we combined a 16SM of gap junction channel gating and rectification with the Hodgkin–Huxley ( HH ) equations [21] . The developed model ( HH-16SM ) allowed us to evaluate the kinetics of junctional conductance during the spread of excitation in neuronal networks . In this study , we replaced 16SM with 36SM for a better evaluation of junctional conductance kinetics . We applied the combined model ( HH-36SM ) to investigate the signal transfer between electrically coupled neurons in response to different types of presynaptic inputs , such as electrotonic signals or action potentials ( APs ) . In this study , we analyse three main aspects of the 36SM with respect to the functional behavior of electrical synapses . Firstly , transjunctional voltage distribution across each channel gate can result to almost instantaneous asymmetric conductance-voltage rectification of gap junction channel . We showed that such rectification of gap junctions can affect the asymmetry of the electrical cell-to-cell signaling , especially the spread of a single AP . Secondly , in the 36SM , the gap junction channel can transit between open and closed states , and probabilities of these transitions depend on voltage across each channel gate . We demonstrated that closing of gap junction channels could be induced by transjunctional voltage spikes , which develop during the spread of excitation . More precisely , our modeling results showed that voltage spikes induced by the trains of APs can cause an accumulation of gap junction conductance decay . As a result , the junctional conductance can significantly decrease in just a few seconds , and substantially modulate electrical signaling between neurons . This short-term plasticity of electrical synapses was supported by our electrophysiological experiments in HeLa cells expressing connexin45 . Thirdly , we suggested that some types of chemical modulation of electrical synapses could be explained by an assumption that the values of 36SM parameters depend on chemical factors . Under such a hypothesis , the chemical modulator would influence the junctional conductance by modifying voltage sensitivity properties of a gap junction channel . In this case , the chemically-induced variation of junctional conductance would be explained by the changed equilibrium of open and closed voltage sensitive channels , and not by a separate chemical gate . To illustrate the feasibility of this idea , we fitted the 36SM to explain the kinetics of connexin36 gap junctional conductance under different concentrations of free magnesium ions ( [Mg2+]i ) . We demonstrated that a long-term ( a few minutes ) plasticity , which is induced by variation in [Mg2+]i , can be adequately reproduced through the changes of 36SM parameters . Thus , the presented model accounts for the complex behavior of electrical synapses under a wide variety of voltage and temporal conditions . Moreover , all these phenomena can be explained by the underlying mechanisms of gap junction channel voltage gating . Such a modeling approach allows one to evaluate the response of neuronal networks , which would be very difficult to measure experimentally .
Junctional conductance of electrical synapses was evaluated using a Markov chain 36-state model of voltage gating , which is detailed in [19] . The model describes the probabilistic behavior of gap junction channels in response to the transjunctional voltage . In the 36SM , the gap junction channel consists of two hemichannels , each enclosing one fast and one slow gates ( Fig 1 ) . Thus , the channel comprises four gates ( fast left , slow left , slow right and fast right ) , all arranged in series ( Fig 1B ) . The fast and the slow gates operate according to a linear kinetic schemes , o↔c and o↔c1↔c2 , respectively ( Fig 1A ) . Thus , gap junction channel can be in 36 ( 2∙3∙3∙2 ) different states , and overall junctional conductance is estimated as an averaged value of each state conductance weighted to their probabilities . Transition probabilities between system states depend on transjunctional voltage distribution across gates , which must be evaluated first . In general , the voltage distribution can be nonlinear due to rectification of unitary conductances of channel gates . The developed HH-36SM combines Hodgkin–Huxley equations that describe excitability of neurons and a 36-state model ( 36SM ) of gap junction channel gating that evaluates conductance of the electrical synapse . More precisely , membrane voltages of the neurons are estimated using the Hodgkin-Huxley model . The resulting transjunctional voltage can affect the junctional conductance , which is evaluated using the 36SM . Thus , the HH-36SM allowed us to simulate electrical signal transfer between neurons connected through modulatable gap junctions .
Asymmetry of electrical synaptic transmission has been observed in numerous studies [41–44] . Such asymmetry might arise due to differences in input resistances ( Rins ) of coupled neurons [29] , even when gap junctions themselves are symmetric . Rin depends on the conductivity of the plasma membrane and its surface area , as well as on the number of neighboring neurons connected through electrical synapses . Another source of electrical synaptic transmission asymmetry is related to instantaneous conductance–voltage rectification of gap junction channel , which results from the inhomogeneous distribution of charged amino acids lining the pore [45] . Such rectification of gap junction channels typically arises in heterotypic junctions under normal conditions [4 , 17 , 46] , but it can also develop in homotypic gap junctions under an asymmetry of intracellular milieu , e . g . gradients of [Mg2+]i [47] . In addition , electrical signaling asymmetry across heterotypic channels can arise with repeated stimulation due to voltage gating ( see Fig 7 ) . This type of asymmetry in electric synaptic transmission is not instantaneous and depends on past history . The other factors that contribute to asymmetry of signaling do not have this property . Our data show that asymmetry in electrotonic cell-to-cell communication is more affected by the difference in Rins of coupled cells ( see Figs 3D and 5D ) , while gap junctional rectification primarily influences an asymmetry of AP transfer between neurons ( Fig 5A–5C ) . This can be explained by the conductance–voltage curves in Fig 2 , which show that conductance changes are small at low voltages ( ±10 mV ) , which typically arise during measurements of coupling coefficients . Significant changes of junctional conductance can only be expressed at high voltages ( ±100 mV ) , which develop during the spread of excitation . We believe that these observations might have practical applications in electrophysiological experiments when studying the strength and rectification properties of electrical synapses . The aforementioned sources of functional asymmetry are independent by nature , e . g . Rin of a neuron directly depends on plasma membrane area , while synaptic rectification is determined by properties of gap junction channels [48] . Thus , they can act antagonistically promoting bidirectionality of electrical synapses , as was demonstrated in the teleost auditory system [4] . Alternatively , if rectification of gap junctions and differences in Rins acted synergistically , it could facilitate unidirectional AP transfer . Thus , unidirectionality , which is a genuine property of chemical synapses , could be executed through electrical synapses alone . Because electrical synaptic transmission is faster than chemical , unidirectional spread of AP through gap junctions might be useful in rapid response warranting behavior such as escape reflex [49 , 50] . Asymmetry of electrical synaptic transmission plays an important role in spike-timing regulation , as was demonstrated in neurons of the thalamic reticular nucleus [44] . In larger networks , even a small asymmetry would add up during the spread of excitation and could significantly affect the latency of AP transfer along neural pathways . This process could be crucial in temporal coding activities , such as coincidence detection , in which gap junctions are reported to play an important role [50] . Presumably , the effect of asymmetry of electrical signaling would be difficult to measure and observe experimentally in highly complex neuronal networks , and a simulation-based approach could provide valuable insights on the role of rectification in network dynamics [51] . It is well established that gap junctional conductance depends on voltage [10] . Our previous [20] and current modeling studies show that decay of junctional conductance can be induced by voltage gating of gap junction channels during bursting activity of neurons . To our knowledge , at least one study reported such spiking activity-dependent reduction of electrical synaptic strength in brain slices [52] . Our data showed that even in gap junctions formed of low-voltage-sensitive Cx36 , this decay exceeds 10% while in more voltage-sensitive Cx isoforms it could reach ~50% over several seconds ( Figs 6 and 7 ) . The magnitude of junctional conductance decrease and duration of its recovery depends not only on Cx properties but also on the firing rates of neurons ( Fig 6 ) . Because the transfer of electrical signal and its asymmetry depends on junctional conductance [53] , an activity-induced inhibition of electrical synapses can significantly diminish ( Fig 6A-b ) or even abolish AP transfer between neurons ( Fig 8C-c ) . Such a role of electrical synaptic plasticity was acknowledged in [54] and was demonstrated by an activity-dependent decrease of junctional conductance together with enhanced asymmetry of electrical synaptic transmission in TRN slices [52] . Heterotypic gap junctions exhibit structure-determined voltage-gating asymmetry , which could result in even more diverse functional behavior with respect to plasticity and directionality than homotypic gap junction . As we showed in Fig 8 , changes in junctional conductance and the response rate of neurons depends on the direction of AP spread with respect to the orientation of heterotypic gap junctions . Thus , heterotypic synapses could promote direction-dependent asymmetry of electrical signal transfer not only by its rectification properties but by asymmetric voltage gating as well . We presume that such processes might have an important functional role in sensory systems where heterotypic electrical synapses are detected [55 , 56] . Regulation of the strength of electrical synapses by a variety of chemical reagents is well established . Others and our data showed that junctional conductance decay caused by chemical uncouplers can be reversed by voltage , while some chemical factors can change voltage sensitivity of Cxs [8 , 38 , 39 , 57] . These observations , as well as the fact that all known chemical uncouplers close gap junction channels fully but not to residual conductance , suggest that some chemical factors act through the slow gate . We implemented this idea by simulating Mg2+-mediated changes in junctional conductance of Cx36 gap junctions using the 36SM . The obtained data revealed that an effect of [Mg2+]i can be relatively well reproduced ( Fig 9 ) assuming variation in values of 36SM parameters , mainly V0 and probabilities of c1↔c2 transitions of slow gates . Moreover , because the voltage sensitivity of the gap junction channels is defined by the same parameters ( see Fig 4 in [20] and Fig 6 in this paper ) , chemically modulated gating would also affect spiking activity-dependent short-term plasticity of the electrical synapse . Our modeling results showed that even a moderate change ( ±20% ) in [Mg2+]i could result in very significant differences in the spread of APs between two neurons ( see Fig 10 ) . Thus , chemically modulated gating of Cx36 can expand the time window of electrical synaptic plasticity for as long as chemical factors are present , which could last for minutes or even hours . Therefore , even Cx36 , which exhibits relatively low voltage sensitivity , could act as a highly modulatable constituent of neuronal networks due to chemically mediated gating . Our modeling results show that a persistent spiking activity or chemical factors could keep a significant proportion of gap junction channels in a closed state . We assume that this process could offer at least a partial explanation to a well-documented ‘low functionality’ of gap junctions , especially those expressed in excitable cells , such as neurons or cardiomyocytes . Low functionality refers to a small fraction of channels residing in the open ( or high conductance ) state . This applies to all connexins , such as Cx36 [58] , Cx43 [26] , Cx45 [57] and Cx57 [59] , examined on this issue , and likely applies to other Cx isoforms . The strength of electrical synapses directly affects the level of synchronization in neuronal networks , which can underlie various physiological processes and pathological brain conditions . For example , increased cortical synchronization correlates with reduced information processing capability in the primary auditory cortex [60] . The rise in junctional conductance can lead to over-synchronization , which is associated with episodes of epileptic seizures . Interestingly , an activity-induced decrease in the coupling of electrical synapses through an intracellular Ca2+ mechanism was observed in the thalamic reticular nucleus of epileptic rats and was proposed to act as a compensatory mechanism to reduce excessive synchronization [61] . Thus , both voltage- and chemically induced gating of gap junction channels can play an important role in shaping activity of neuronal networks through modulation of neuronal synchrony . In addition , short-term plasticity induced through voltage gating of electrical synapses could contribute to lateral inhibition and resulting center-surround effect , which is important in sensory systems of the CNS . This hypothesis is supported by studies showing that more voltage-sensitive Cx isoforms are expressed in the structures associated with sensory functions . For example , one of the most voltage-sensitive Cxs , mouse Cx57 and its human homolog Cx62 are expressed in horizontal cells of the retina [62] , while Cx45 , which is significantly more voltage-sensitive than Cx36 , predominates in the olfactory bulb [34] . The chemically mediated gating could play an important role in regulating longer term changes , especially in less-voltage-sensitive Cx36 . For example , it was reported that Cx36 plays an important role in shifting between sleep and wake states [63] . We believe that the unique sensitivity of Cx36 to Mg2+ could contribute to this process . This view is supported by accompanying changes in ATP levels , which effectively influence [Mg2+]i . It was reported that ATP levels increase during the initial hours of sleep in wake-active regions of rat brain [64] . This should decrease [Mg2+]i and , consequently , increase conductance of Cx36 gap junctions . As a result , an increased synchronization could suppress activities in brain regions associated with the waking state , thus maintaining sleep . In this study , we used the Hodgkin–Huxley equations to describe excitability of neurons . The developed model can be adapted to various brain regions and circuits by choosing an appropriate set of ionic currents . For example , the inclusion of Ca2+ currents , which underlie bursting trains of APs in thalamic relay neurons [65] , might be relevant for short-term plasticity as well as for chemical modulation of electrical synapses . Furthermore , major principles used to develop an HH-36SM can be applied in cardiac tissue modeling , provided that the Hodgkin–Huxley equations are replaced by those specific for cardiomyocytes [66 , 67] . Cardiomyocytes are predominantly connected through Cx43 , Cx40 and Cx45 , which are more voltage sensitive than Cx36; therefore , it might exhibit more expressed activity-dependent conductance decrease , especially during tachyarrhythmias . Furthermore , chemically mediated gating of cardiac gap junction channels , e . g . by acidification [68] , could be important in describing enhanced arrhythmogenicity of the ischemic myocardium [69] . Obviously , the 36SM of gap junction channel voltage gating is a simplification of complex processes underlying changes of electrical synaptic strength . However , we believe that rectification and voltage gating properties of gap junction channel can be reasonably well reproduced using the 36SM . On the other hand , an inclusion of chemical modulation into 36SM is far less explored . So far we made only the first steps in this direction to explain Cx36 mediation by [Mg2+]i , and presented modeling results ( Fig 10 ) are obtained from just a few data points . Moreover , cytosolic conditions are rarely defined by a single chemical factor , and various different reagents might affect electrical synapses synergistically or antagonistically . For example , our preliminary data suggest that [Mg2+]i effect on Cx36 gap junctions might depend on the pH level . In addition , modulation of electrical synapses by other chemical reagents , such as Ca2+ ions , might be more relevant for the spread of excitation than that of [Mg2+]i . In this study , we simulated electrical synaptic transmission between two cells connected through a soma-somatic gap junction . For a more realistic neuronal network simulation , it would be beneficial to include dendro-dendritic connections , which are far more prevalent in mammalian brain . Another important extension of our model would be an inclusion of chemical synapses . Presumably , this would allow one to study an interaction between chemical and electrical synapses , which was observed in numerous experimental studies [70] . However , all physiologically relevant extensions , and especially an increased number of cells and synapses , might require a large amount of computational recourses . To our knowledge , Hodgkin-Huxley type models are rarely applied for large neuronal network simulation due to computation time constraints . This problem would be enhanced by our modeling approach , because evaluation of junctional conductance using the 36SM consumes ~95 percent of overall computation time . We presume that simulation time could be decreased by two different approaches: 1 ) Creation of a more simplistic model of gap junction voltage gating , which would roughly describe relative changes of junctional conductance in response to a single AP . Somewhat similar approach is applied in mathematical models of chemical synapses [71] . This would allow one to combine a model of electrical synapse with integrate-and-fire type models , which are often used for simulation of large neuronal networks . 2 ) Application of advanced computation techniques , such as an extensive parallelization together with graphic processing unit computation . | In most computational models of neuronal networks , it is assumed that electrical synapses have a constant and ohmic conductance . However , numerous experimental studies demonstrate that connexin-based channels expressed in neuronal gap junctions can change their conductance in response to a transjunctional voltage or various chemical reagents . In addition , electrical synapses may exhibit direction-dependent asymmetry of signal transfer . To account for all these phenomena , we combined a 36-state model of gap junction channel gating with Hodgkin–Huxley equations , which describes neuronal excitability . The combined model ( HH-36SM ) allowed us to evaluate the kinetics of junctional conductance during the spread of electrical signal or in response to chemical factors . Our modeling results , which were based on experimental data , demonstrated that electrical synapses exhibit a complex behavior that can strongly affect the response of coupled neurons . We suggest that the proposed modeling approach is also applicable to describe the behavior of cardiac or other excitable cell networks interconnected through gap junction channels . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Discussion"
] | [
"cell",
"physiology",
"ion",
"channel",
"gating",
"medicine",
"and",
"health",
"sciences",
"action",
"potentials",
"neural",
"networks",
"nervous",
"system",
"membrane",
"potential",
"junctional",
"complexes",
"electrophysiology",
"neuroscience",
"gap",
"junctions",
"ion",
"channels",
"synaptic",
"plasticity",
"electrical",
"synapses",
"neuronal",
"plasticity",
"developmental",
"neuroscience",
"computer",
"and",
"information",
"sciences",
"animal",
"cells",
"proteins",
"biophysics",
"physics",
"biochemistry",
"cellular",
"neuroscience",
"cell",
"biology",
"anatomy",
"synapses",
"physiology",
"neurons",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"neurophysiology"
] | 2017 | Functional asymmetry and plasticity of electrical synapses interconnecting neurons through a 36-state model of gap junction channel gating |
The 2014/2015 West African Ebola Virus Disease ( EVD ) outbreak attracted global attention . Numerous opinions claimed that the global response was impaired , in part because , the EVD research was neglected , although quantitative or qualitative studies did not exist . Our objective was to analyse how the EVD research landscape evolved by exploring the existing research network and its communities before and during the outbreak in West Africa . Social network analysis ( SNA ) was used to analyse collaborations between institutions named by co-authors as affiliations in publications on EVD . Bibliometric data of publications on EVD between 1976 and 2015 was collected from Thomson Reuters’ Web of Science Core Collection ( WoS ) . Freely available software was used for network analysis at a global-level and for 10-year periods . The networks are presented as undirected-weighted graphs . Rankings by degree and betweenness were calculated to identify central and powerful network positions; modularity function was used to identify research communities . Overall 4 , 587 publications were identified , of which 2 , 528 were original research articles . Those yielded 1 , 644 authors’ affiliated institutions and 9 , 907 connections for co-authorship network construction . The majority of institutions were from the USA , Canada and Europe . Collaborations with research partners on the African continent did exist , but less frequently . Around six highly connected organisations in the network were identified with powerful and broker positions . Network characteristics varied widely among the 10-year periods and evolved from 30 to 1 , 489 institutions and 60 to 9 , 176 connections respectively . Most influential actors are from public or governmental institutions whereas private sector actors , in particular the pharmaceutical industry , are largely absent . Research output on EVD has increased over time and surged during the 2014/2015 outbreak . The overall EVD research network is organised around a few key actors , signalling a concentration of expertise but leaving room for increased cooperation with other institutions especially from affected countries . Finding innovative ways to maintain support for these pivotal actors while steering the global EVD research network towards an agenda driven by agreed , prioritized needs and finding ways to better integrate currently peripheral and newer expertise may accelerate the translation of research into the development of necessary live saving products for EVD ahead of the next outbreak .
The 2014/2015 West African Ebola Virus Disease ( EVD ) outbreak with more than 28 , 000 cases and 11 , 000 deaths , was a public health emergency of international concern [1 , 2] . Although EVD was discovered in the former Zaire ( now: Democratic Republic of Congo ) more than 40 years ago , the absence of treatment generated global alarm and raised questions on the state of EVD research . Studies analysing EVD transmission and clinical trials testing EVD treatments or vaccines have been difficult due to the small number of infected cases in previous outbreaks [3 , 4] . Moreover , the pharmaceutical industry has been criticized for neglecting EVD research because it is not profitable enough as EVD occurred rarely and mostly in impoverished African communities [3 , 5–7] . EVD outbreaks have attracted general public attention since the mid-90s , benefitting science funding , leading to increased publications , but EVD research funding is mostly spent outside of affected African countries and research capacity building there was neglected [8] . The World Health Organization ( WHO ) called for greater transparency and better sharing of results from clinical trials as being a necessary contribution to facilitate research and development ( R&D ) for the benefit of science and patients [9] and published a research priority agenda [10] . The necessity for increased transparency also applies to any existing EVD research and expertise to improve the value and efficiency of research efforts . In order to enhance the understanding of on-going EVD research activities and its communities , social network analysis ( SNA ) of bibliometric data of EVD related scientific publications can be used . Since co-authorships are the most visible and accessible indicator for collaborations , co-authorship-based SNA studies can be used to measure the presence of research collaborations and their evolution over time [11–13] . SNA metrics can reveal network patterns and identify its most central and influential actors [14–16] . The volume of publications , in combination with results from a co-authorship network analysis , can serve as a proxy indicator for R&D . Besides mapping the research landscape [17] , especially co-authorship network analysis can provide insight into the degree of research governance and be relevant for strategic research planning [18 , 19] . Moreover , information from collaboration networks can be used to identify potential collaborations in order to improve research communication and therefore maybe also influence research outcomes [12 , 20] . The aim of this study is to identify EVD research activities and to analyse the structure of the evolving EVD research community network over time to map existing research collaborations and influential actors based on centrality network metrics .
Bibliometrics of 2 , 528 articles resulting for our WoS search were exported as tab-delimited data and imported into MS Excel as one bibliometric data set ( Fig 2 ) . In the raw data set each entry referred to one publication . We included data on title , authors , address of authors’ affiliated institution , publication year , source , language , document type , cited references , funding agency , publisher and subject category in further analysis . Other columns were deleted from the data set . Information on addresses of author’s affiliated institution , e . g . institution name , sub-departments and institution address including city and country , were split into separate columns . Data processing and further cleaning was performed using the software AppleScript [22] and OpenRefine [23] . Name disambiguation , e . g . Centers for Disease Control and Prevention was abbreviated as CDC , Ctr Dis Contr and Centers Dis Cont , orders within names , e . g . Univ Washington and Washington Univ or name spellings , e . g . , Univ Georgia , UNIV GEORGIA were identified and harmonised using OpenRefine algorithms or manually . Missing data , e . g . missing country information of an affiliated institution , were substituted by manual web search . If an institution name appeared with addresses in different locations in the data set , e . g . WHO with location Switzerland and location Copenhagen e . g . due to different regional offices , different locations were considered for construction of the network to account for institutions international representations . Institutions duplicates originating from publications with multiple co-authors affiliated with the same institutions were eliminated to ensure a single weighting of institutions . The free online application Table2net was used to extract network information from the refined data set to construct a Gephi readable file [24] . Network nodes ( i . e . actors ) are institutions named as authors’ affiliations in original research publications . Network edges are titles of joint publications from authors’ affiliated institutions . The free software Gephi was used to calculate network metrics and visualise the networks [25] . Network analysis provides various tools and metrics in order to assess different notions of importance of individual nodes and node groups . As the simplest metric of centrality we calculated each node's degree , as the sum of direct links to other nodes . Nodes with more direct connections are considered more central . The average node degree captures the number of actors that each actor is connected with on average . The average weighted node degree also takes the weight of a connection between a pair of nodes into account [26 , 27] . Betweenness centrality measures the frequency with which a particular node lays on the shortest paths between all other node pairs . Therefore , nodes with a high betweenness are considered to have a broker position as they connect many other nodes and thus have a large influence on the transfer of items through the network , under the assumption that item transfer follows the shortest paths [26 , 28] . We used a betweenness calculation algorithm for weighted graphs as developed by Opsahl [29] . Besides positional properties of the nodes within the network , metrics are capturing topological aspects of the network as a whole . This information can provide an insight on the evolution at network level . Density measures were calculated to assess the connectivity of the network . The density of a network is defined as the total number of existing edges divided by the total number of possible connections . If edges exist between all nodes ( density = 1 ) a network is considered completely dense [26 , 28] . Since density captures the probable feasible number of connections in a network , it is an indicator for possible community building [30] or innovation flow within a network [15] . Communities within the network were detected using Gephi’s modularity algorithm . Modularity measures the degree of separation of a network into modules or clusters ( communities ) . While a modularity value of 1 indicates that the actors separate perfectly into self-contained clusters , a value of - . 5 suggest the opposite , a homogeneously connected network [27 , 31] . Networks with a high modularity score employ dense connections between nodes within the modules but sparse connections between nodes from different modules . For visual presentation of network metric calculations we used Gephi's Force Atlas II algorithm in log-linear mode optimized towards hub dissuasion [32] .
Systematic search in WoS for publications containing “Ebola*” yielded a total of 4 , 587 publications between 1976 and 2015 , including original articles ( 2 , 531 ) , editorial material ( 659 ) , news items ( 437 ) , reviews ( 415 ) , letters ( 325 ) , meeting abstracts ( 157 ) , corrections ( 36 ) , notes ( 14 ) , reprints ( 7 ) , biographical items ( 4 ) and book reviews ( 2 ) . Amongst the 2 , 531 original articles were 75 article proceedings and five article book chapters . Three of those publications appeared with anonymous authors and were therefore deleted for social network analysis ( Figs 1 & 2 ) . The first EVD research article was published in 1977 , shortly after the first noted EVD outbreak in 1976 . Only few EVD publications were visible until the early nineties , whereas from 1994 onwards the number of yearly EVD publications increased continuously ( Fig 3 ) . Since 1994 a higher frequency of EVD outbreaks were recorded and more EVD cases were being detected in almost every year . Several localised EVD outbreaks in Africa have occurred with up to several hundred cases . The initial EVD outbreak in 1976 , with a relatively high number of reported cases ( >600 ) , was followed by only a small number of publications on EVD research . No EVD outbreaks were reported between 1979 and 1994 and hardly any publications were published on the topic . The number of publications increased gradually and continuously after the second outbreak in 1994 , although compared to the 1976 outbreak only about one-tenth of cases were reported ( Fig 4 ) . A substantial increase in EVD research publications occurred during the 2014/2015 West African outbreak . An almost 10-fold increase from 2013 ( 171 ) , 2014 ( 772 ) to 2015 ( 1 , 621 ) was visible for almost all document types , but it was most pronounced for editorials ( 5 , 220 , 343 ) , letters ( 1 , 75 , 213 ) , news items ( 4 , 190 , 118 ) and meeting abstracts ( 9 , 5 , 66 ) respectively . An increase in reprints , notes , biographical items and book reviews was not detected . Bibliometrics of 2 , 528 original research articles were used for social network analysis . Based on their co-authors’ affiliated institutions a global network including institutions from 101 different countries with 704 connections was constructed ( Figs 5 & 6 ) . Research institutions in the United States ( US ) are among the most highly connected institutions in EVD research ( degree ( d ) = 80 ) . They are mostly connected to institutions in Canada ( d = 40 ) with an edge weight ( ew ) of 130 and Europe , especially Germany ( d = 53 , ew = 110 ) , the United Kingdom ( UK ) ( d = 60 , ew = 90 ) and France ( d = 57 , ew = 51 ) , but also to Japan ( d = 32 , ew = 99 ) . Connections between US institutions and institutions in EVD affected African countries are less frequent ( e . g . Guinea-USA ew = 14 , Sierra Leone-USA ew = 32 , Liberia-USA ew = 30 ) . However , institutions in Sierra Leone and Guinea ( both d = 32 ) and other African countries , especially Nigeria , Uganda and Ghana , are embedded in the global research network with connections to UK , Germany , France and Switzerland . The overall density of the global country-level EVD research network measures 0 . 15 , with an average degree of 14 . 65 and an average weighted degree of 61 . 01 . Amongst all collaborations on country-level , nine research communities were identified using modularity-based community detection and visualised by different colours ( Fig 6 ) . The largest community ( red ) is centred around the US with strong collaborations to Canada , Germany and the UK , representing 59 . 41% of the co-authorships collaborations ( weighted edges ) . Another large community is a ( mostly francophone ) European–African community ( blue ) representing 31 . 68% of all co-authorships connections . Among all published original research articles between 1976 and 2015 a total of 1 , 644 co-author's affiliated institutions were named , which yielded 9 , 907 co-authorship connections in the overall research network ( Fig 7 ) . The main actors according to degree are the US government ( CDC USA , d = 353; NIH , d = 315; USAMRIID , d = 283 ) and WHO ( d = 256 ) . Other prominent actors are from the US and European countries . Most central institutions are publicly funded ( e . g . CDC USA , USAMRIID ) , government research institutions ( e . g . BNI , ISERM ) , ( mostly public ) universities ( e . g . Uni London , Univ Marburg ) or international institutions ( e . g . WHO ) or non-governmental institutions ( NGOs ) ( e . g . MSF ) . Modularity analysis reveals 166 communities within the network ( Fig 7 ) , whereas the largest community ( blue ) represents 17 . 33% of the total network nodes and the second largest ( green ) represents 14 . 44% of the network nodes . Numerous smaller and less connected communities exist in the periphery , with some being entirely disconnected from the main network . The temporal development of the research network is visualised over four 10-year time periods ( Figs 8 , 9 , 10 & 11 ) . In the first decade 1976–1985 , ( Fig 8 ) the network consists of only a few actors , with one large central cluster surrounded by four smaller clusters . The German Bernhard-Nocht Institute ( BNI ) has the highest centrality degree ( d = 11 ) , closely followed by the Institut Pasteur , PHLS Center for Microbiology and Research ( Salisbury , UK ) and USAMRIID . The CDC USA is a central institution ( d = 7 ) of a smaller cluster , publishing with African partners ( Kenyan Ministry of Health ) others . Smaller research groups in Kenya ( Kemri Wellcome Trust , Institute of Primate Research , Kenya Trypanosomiasis Research Institute ) , UK and US published together , but had no connections with others . In the second decade 1986–1995 , ( Fig 9 ) two larger , but separate , research communities evolved . One francophone French-Swiss-African community with a homogenous structure in which the Institut Pasteur published mainly with the University of Basel , Institut de recherche pour le développement ( IRD ) , Ecole national veterinaire Lyon and the Hospital Bichat Claude Bernard Paris . The other community consists mostly of American and German institutions , with three main actors ( USAMRIID , CDC USA and the University of Marburg ) , where the USAMRIID and CDC USA connect this community . During this period the WHO had its first appearance as a disconnected actor . All institutions in the network of the second decade are public entities . With the occurrence of new EVD outbreaks in 1994/1995 the EVD research network grew in the third decade 1996–2005 , ( Fig 10 ) into a star-like structure with surrounding chains . During this decade the CDC USA evolved as the most central actor ( d = 87 ) . The University of Marburg ( d = 54 ) , USAMRIID ( d = 52 ) , WHO ( d = 46 ) and NIH ( d = 36 ) remain central but less prominent actors . The network of the fourth decade 2006–2015 , ( Fig 11 ) is skewed by publications in 2014/2015 . During this time only few public research institutions and university actors dominate the research collaborations but numerous new actors appeared . Prominent cooperation exist between CDC USA and WHO and CDC , NIH and USAMRIID . While the transnational WHO was well embedded in the network over these last two decades , all main network actors are public institutions , mostly from the US and European countries . While the global EVD research network remains relatively consistent in the first two decades , the third and in particular the forth decade shows substantial overall increase in the number of institutions and the links between them ( Table 1 ) . Simultaneously the average node degree and weighted node degree increased over time , which indicates a growing number of collaborations and research activity per institution . The decreasing density of the network over all decades indicates a decreasing number of realised edges between nodes relative to the total number of possible edges . The increasing average node degree implies a growing number of research connections per institution . The number of communities increased in line with number of nodes . The high modularity values show that the solutions of the community detection algorithm reflect the substructures of the graph well , i . e . the increase in communities is unlikely to represent a sheer increase in volume , but rather seems to capture an evolution of the field of EVD into several smaller communities . A degree distribution analysis of the EVD research network in the fourth decade shows a skewed node-degree distribution ( Fig 12 ) . While almost 100 nodes appear with a degree of zero ( d = 0 ) , indicating no collaboration at all , only few institutions have a very high degree above 160 ( mean 12 . 24; median 5 ) . Most institutions had a degree of less than five ( d≤5 ) as they were named as affiliations by authors of few publications by authors that published with only few co-authors . The few very well connected institutions , such as NIH and CDC USA , are the key actors in this period . In fact the CDC USA has maintained a very central position in the network over all time periods . The private NGO Médecins Sans Frontières ( MSF ) has only recently emerged within the network and is centrally embedded with a high degree ( d = 157 ) .
Since the first reported EVD outbreak in 1976 until today the total number of publications on EVD in WoS has exceeded more than 4500 publications , of which 2528 were original research articles . Like in scientometric analyses we used joint publishing as a proxy indicator of scientific collaboration [17] and thus knowledge exchange for our SNA of the co-authorship network [11 , 13 , 30] . Indeed for the EVD overall network we identified research contributions from 1 , 644 research institutions in 101 countries; most actors are indeed coming from the US [17] . Since 1994 EVD research publications have increased continuously , steadily and independently of the major West African outbreak . This growth in publications is mirrored by a growth in the number of institutions ( from 30 to 1 , 489 ) and edges ( from 60 to 9 , 176 ) and therefore on-going network growth accompanied by a decreasing network density . The overall network is an extensive aggregation of 166 different communities with a clearly dominant anglophone and francophone community . This same dominance is seen when analysing the most central actors by degree and betweenness centrality both confirming the dominance of 10 institutions in powerful , control or broker positions in the network [11 , 33 , 34] . The pattern of a growing EVD network in size but with a reducing density is characterised by some outliers ( 106 institutions not connected ) , frequently less connected contributions from developing countries and the private sector , but with a strong and stable core of dominant or ‘central’ institutions . These characteristics of the network are supported by many of the analyses we performed . For example the relatively and increasingly poorly connected nature of the network ( network density ) , the heavily skewed node degree distribution with the median node degree remaining rather constant , the relatively compact nature of the network ( path lengths ) and the strong centralisation showing a dominance of a few very strongly connected actors and many poorly connected actors . Although we acknowledge that our analysis is weakened by the absence of a comparator network ( a common challenge in emerging research fields ) , we also believe that our analysis bring some added value . For example SNA metrics for the overall network shows a density of 0 . 007 and calculating network density for each decade individually showed progressively decreasing density from 0 . 138 in the first decade to 0 . 008 in the last decade . While this is largely influenced by both the size ( the more actors a network includes , the more difficult it is for all actors to be connected ) and also the correspondingly rapid growth in the network ( connections take time to build ) , we still believe that these figures should raise questions about whether the network–and therefore research outputs–could benefit from greater connectivity and linkages and in doing so greater optimise knowledge transfer and the spread of innovation [15] . The node degree distribution ( for the last decade from 2006–2015 ) further confirms both the observed increase in the average node degree is attributable to only a few central actors whereas the overall network was not well connected in this period . Thus , the network growth during the 2014/2015 epidemic diluted connectivity , at a time when collaboration was arguably most needed . These observations are built on when we look further at the node degree distribution for 2006–2015 . This confirms that while most actors only had few connections during this time , some actors are extremely connected . This distribution form has been described as “power law” or “scale free distribution” and is typically observed amongst poorly connected networks [35 , 36] . This ‘concentrated core’ is corroborated by the high number of the average weighted node degree ( 17 . 89 ) , in contrast to the average node degree ( 12 . 05 ) , which is also an indicator that some actors in the EVD network are connected more strongly to each other than others due to repeated publishing [27] . It shows that these actors have on-going collaborations , share research results intensely by jointly publishing—but focus sharing amongst their co-authors . This latter finding is something confirmed by our SNA results , which show strong centralisation amongst six institutions ( CDC USA , NIH , USAMRIID , WHO , the University of Marburg and the University of Harvard ) , suggesting that knowledge is mostly exchanged within the network between and/or through these actors . Centrality is a measure of power in SNA [37] , this is especially the case for our central actors whose knowledge broker status is confirmed with regard to EVD research due to their high degree and betweenness centralities . Additionally , observation of the path lengths reveal further insight into the efficiency of information exchange , with the shorter the average path length of a network diameter , the more efficient is information exchanged within the network structure [26 , 35] . We found that the average paths lengths ( 3 . 02 ) of the overall network is lower than the average node degree ( 12 . 05 ) , indicating both that some institutions have a lot of direct neighbours and that on average nodes can reach other nodes by crossing only two other nodes . The network diameter ( 8 . 0 ) suggests that sub-graphs within the network do not span more than across a chain of eight nodes . Taking both aspects into account this implies that the overall structure of the network is characterized by isolated and weakly connected components , i . e . localized small networks that have only few relations amongst each other . Although our study cannot , unfortunately , reveal anything about the ‘type’ of research conducted , observations on the type of research institution maybe serve as a proxy for this insight . Two notable observations here were both the relative underrepresentation and disconnectedness in the overall network of both research institutions from affected countries and the private sector . Among the unconnected nodes appear some private industry actors ( e . g . , Novartis Vaccines , Biohelix Corp , Baxter Bioscience and Oravax Inc . ) , in addition to African universities such as the University of Benin and the University of Mbarara . While there may be many good reasons that explain the disconnectedness , for example proprietary restrictions to collaboration ( in the case of industry ) , new entrants to the field or for resource-related barriers to International collaboration . This observation remains significant for a number of reasons , presumably both of these actor types posses’ unique and distinct knowledge and capabilities that could diversify and strengthen the expertise within the network if better and more broadly integrated , this is likely even more the case during a public health emergency of international concern . Also , this ability to identify disconnected but valuable nodes , demonstrates a great added value of tools such as SNA . Finally the recent entry into the network of non-traditional research actors such as MSF should be welcomed , especially as endemic country capacity is being developed and integrated into international networks , due to their unique position as being close to patients in the field yet able to advocate–distant funders–on the need for a well-supported , needs-driven research agenda [5 , 38] . We believe the structure , nature and evolution of the international EVD research network described in this paper presents some learnings for policy . Looking positively , the network itself has maintained a similar structure–a relatively compact network with a few consistent actors at its core–over the four decades studied , implying it is a stable constellation . This institutional memory provides a solid foundation for knowledge maintenance over time , indeed without central actors networks might be disrupted and knowledge exchange hampered [30] . The growth in the network over time through the entry of new actors , particularly since 2014/2015 , is positive as it likely indicates the arrival of new ideas and approaches . However although collaboration has increased over time , our analysis found that the network remains relatively poorly connected . Hence there may be an additional role for the ‘central actors’ to expand their role beyond a hub for dissemination and exchange into a facilitator for integrating the newer actors and expertise into the network . Additional opportunities presented by the network analysis include: a reflection on the , perhaps , over-reliance or vulnerability to the network of all of the ‘central actors’ being public government or university institutions . The importance of predictable , sustainable , funding flows to their continued role as network ‘brokers’ feels more exposed in these current financially and politically turbulent times . While the dominance of these institutions is not surprising , we assume that they have the infrastructure , capability and public-financing , it may represent a weakness in two respects: firstly , with respect to its insufficiently diverse expertise mix , particularly with respect to the translation of this research into the development of tangible , context-relevant tools and capacity building in affected countries [8 , 39]; secondly , with respect to the risk of over-centralising expertise , resulting in the stifling or suppression of innovation and growth and development of new ideas . Finally , in small research areas for diseases predominantly impacting the lives of those in low-income countries such as EVD , the inherent market failures indicate that this reliance of public-financing will likely continue [Wölfel in: 3 , 5–7] . Given this , we believe , that a valuable insight from our study is to observe ways in which the network efficiency could be enhanced to extract greater patient-impact from the public financing inputs . For example: focused efforts on integrating new collaborators into the network , provision of tools to enhance the productivity and improved transparency and sharing of research data [9 , 40] the identification of expertise gaps and targeted filling of these gaps and lastly , but perhaps most importantly , National alignment , focus and financing coordination ( strategic research planning ) around the globally agreed prioritised research agenda [41] . Although many of these calls have already been made by many actors , particularly since the 2014/2015 EVD outbreak we believe this study represents an important empirical tool to support these calls and inform National and global policy development as the global community works to avert the next EVD outbreak . The use of bibliometric data has intrinsic limitations and restrictions related to any analysis of secondary data and where data ceases to provide information , in particular in relation to content or results of published research . Two major limitations to our study were identified and previously highlighted . The first being the absence of other publications with which to contextualise and compare our results . This absence of relativity in our conclusions limits the comparative value of our findings although the absolute data remain valid . Although SNA is increasingly being used as a tool to analyses research areas it remains a relatively new field so we are optimistic that this is a time-limited constraint . Secondly , we acknowledge that our study would be greatly enriched by an ability to analyse the data by ‘type’ of research not only type of publication i . e . basic , applied , clinical , implementation research , translation , health systems etc . However , at present , this is not a search field within WoS , so we were unable to attain the source data . Should key , public , medical , search engines enable this in the future , SNA such as ours would be an even more powerful tool to provide insight into research focus and productivity . This analysis we believe would have great value–supplementing existing financing and development pipeline analyses [42 , 43]—in providing a more granular understanding of product development gaps and the persistent absence of tools for the prevention , diagnosis and treatment of EVD [6 , 44] . Our analysis of decreasing network density over time could have been further triangulated with the use of an additional metric such as the percentage of the giant component or the clustering coefficient . Other limitations include reporting delays and the possibility that some publications were not included in the WoS database , however sample testing of other databases , including PubMed . gov , did not reveal other publications on EVD . Although the impact of missing publications was likely small future studies could aggregate studies from diverse databases and in particular try to assess contribution of private industries R&D . Despite manual and automated attempts to resolve challenges with institution name cleaning and disambiguation it cannot be excluded that some actors and/or relationships were not captured or were captured incorrectly . Although unlikely , errors of the software used cannot be completely excluded and different algorithms might lead to different presentations of results . Therefore network visualisations should be critically assessed in context to minimise misinterpretations . We further note that GeoLayout visualisation can be misleading since it locates the African continent in the map centre and visualised edges may overlap nodes . For this reason a country distribution was processed additionally with Force Atlas 2 . The use of only free available software and easy accessible bibliometric data from WoS both facilitate the easy reproducibility of our study . We conducted the first systematic landscaping of published EVD global research bibliometrics using SNA tools for analysis and visualisation . Since 1976 Ebola outbreak EVD research , numbers of authors and affiliated institutions and links between them are constantly increasing , mostly independent from outbreaks and in-particular in the past two decades . The overall EVD research network is organised around a few co-authoring key actors , mostly publicly financed . Low network density indicates room for increased cooperation between institutions , in-particular links to less connected and more peripheral institutions could foster knowledge exchange and innovation . Key network actors , such as the CDC USA , maintained network coherence over time–and probably kept EVD research on-going . Limited scientific collaboration of research organisations from LMIC and the private industry , and how they utilise their expertise and knowledge , is neglected . However , the absence of effective treatments for EVD questions the existing EVD research network efficacy and efficiency and suggests the need for both direction and structure to optimize the network to focus on research relevant for treatments . Since most institutions in the global network are publicly funded , guidance to direct and re-orientate research might be facilitated by funders ( through calls targeting knowledge and translation gaps ) and be offered by supranational policy setting entities such as WHO and its Global Observatory on Health Research and Development . Further in-depth quantitative and qualitative analysis , e . g . text mining of publications abstracts , analysis of EVD research study methods and separate R&D product pipeline analysis , is recommended to ensure empirically based strategic research guidance and relevant to EVD product development . In any case , SNA of co-authorship networks is an innovative tool to evaluate research collaborations between individuals , organizations and countries , contributes to the understanding of the evolution of research networks and should be used for strategic research planning and a regular monitoring . | Ebola Virus Disease ( EVD ) research publications were used to analyse and visualise collaborations between institutions jointly publishing research results , using freely available social network analysis tools . Constructed co-authorship networks between author affiliated institutions showed EVD research publications increased and networks evolved over time . The global network is organised around a few co-authoring , mostly publicly financed key actors , highly connected with powerful and broker positions . The results present an extensive narrative how modern empirical scientific methods for data processing and translation can supplement evidence-based arguments for public discussion on the status and focus of global EVD research . Based on the network characteristics or concentration of expertise , we recommend a globally agreed and prioritized EVD research agenda may facilitate the translation of this research into new EVD tools . Also , to analyse research networks regularly to enable public discussion on the direction in which research could be organized and optimised . We would like to encourage others to utilize our methods with open access tools to enhance new methods to the field of NTD R&D . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"united",
"states",
"bibliometrics",
"sociology",
"geographical",
"locations",
"social",
"sciences",
"north",
"america",
"network",
"analysis",
"social",
"networks",
"information",
"technology",
"data",
"processing",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"economics",
"research",
"assessment",
"centrality",
"people",
"and",
"places",
"finance",
"social",
"research"
] | 2017 | Analysing published global Ebola Virus Disease research using social network analysis |
Insect vitellogenin ( Vg ) has been considered to be synthesized in the fat body . Here , we found that abundant Vg protein is synthesized in Laodelphax striatellus hemocytes as well . We also determined that only the hemocyte-produced Vg binds to Rice stripe virus ( RSV ) in vivo . Examination of the subunit composition of L . striatellus Vg ( LsVg ) revealed that LsVg was processed differently after its expression in different tissues . The LsVg subunit able to bind to RSV exist stably only in hemocytes , while fat body-produced LsVg lacks the RSV-interacting subunit . Nymph and male L . striatellus individuals also synthesize Vg but only in hemocytes , and the proteins co-localize with RSV . We observed that knockdown of LsVg transcripts by RNA interference decreased the RSV titer in the hemolymph , and thus interfered with systemic virus infection . Our results reveal the sex-independent expression and tissue-specific processing of LsVg and also unprecedentedly connect the function of this protein in mediating virus transmission to its particular molecular forms existing in tissues previously known as non-Vg producing .
Rice stripe disease is a serious problem during rice production , with epidemics occurring repeatedly in China , Japan and Korea [1–3] . Transmission of the causative pathogen , Rice stripe virus ( RSV ) , is completely dependent on insect vectors , the most important of which is the small brown planthopper ( SBPH; Laodelphax striatellus ) [4] . RSV is transmitted by L . striatellus in a persistent-propagative manner [4] . The RSV filamentous ribonucleoprotein particles ( RNPs ) are ingested by L . striatellus individuals feeding on RSV-infected plants . Once inside the insect , the virus invades the midgut epithelium to establish infection; it then spreads within the gut and disseminates into the hemolymph . From the hemolymph , the virus further infects various L . striatellus tissues , including the salivary glands . RSV is then horizontally transmitted from the salivary glands into a healthy plant , and also invades the female ovaries , from where it is vertically transmitted to the offspring [5 , 6] . Vertical transmission results in naturally existing RSV-infected L . striatellus , which presents a further challenge in disease control . RSV RNPs contain four single-stranded RNAs , and the major nucleocapsid protein ( CP ) encoded by the ORF at the 5’ half of the viral complementary RNA3 [4] . Thus , CP is considered the key viral component for specifically interacting with the vector components and plays roles in RSV transmission . Recently we demonstrated that CP interacts with L . striatellus vitellogenin ( Vg ) in vitro , and that the virions co-localize with Vg in the insect germarium [5] . RSV accomplishes its vertical transmission by binding to Vg and via the uptake of Vg by developing oocytes [5] . In the present study , we focused on the molecular events of the Vg–RSV interaction prior to oocyte penetration . Detailed knowledge of virus transmission mechanisms is required for the design of novel disease control strategies . Vgs are precursors of the major egg storage proteins in many oviparous animals [7 , 8] . Insect Vg is usually synthesized extra-ovarially by the fat body . After processing and modification , also in the fat body , the protein is secreted into the hemolymph and taken up by oocytes via receptor-mediated endocytosis [9–12] . All the known Vgs have been reported to possess modifications; however , the extent of Vg modifications varies considerably . Vg undergoes co- and post-translational modifications , as well as proteolytic cleavage [7 , 12–14] . The primary Vg precursor separates into Vg subunits upon proteolytic cleavage . All insect Vgs , excluding those of the honeybee suborder Apocrita , are cleaved in vivo at the tetra-residue motif R-X-R/K-R by subtilisin-like endoproteases [8 , 15–18] . This conserved R-X-R/K-R motif is located near the N-terminus and is flanked by polyserine tracts ( see review [7] ) . Cleavage at this motif gives rise to two subunits , one large ( 140–190 kDa ) and one small ( 40–60 kDa ) ( see review [19] ) . In some insects , the Vg precursor contains additional RXXR motifs , and the large subunit is further cleaved into two medium-sized ( ~90–110 kDa ) polypeptides [17 , 20–22] . Vg subunits produced by proteolytic cleavage are usually assembled in tissues and secreted into the hemolymph as oligomeric proteins . However , only the small subunit is secreted into the hemolymph , while the large subunit is consumed in the fat body [23] . Vg was initially regarded as a female-specific protein; however , Vg synthesis , albeit in small quantity , has been shown to occur in males and sexually immature animals , indicating that the function of Vg extends beyond serving as an energy reserve for the nourishment of developing embryos [14 , 24 , 25] . In recent years , accumulating data have shown that Vg from fish plays a role in immune responses [26–32] , either as a pattern recognition molecule to recognize bacteria , or as an opsonin to enhance macrophage phagocytosis [27 , 33] . Moreover , Vg has been reported to directly kill bacteria via interaction with lipopolysaccharides and lipoteichoic acid present in bacterial cell walls [28] , and to neutralize viruses by binding to and creating cross-links between virions [26] . A few studies on the immunological properties of Vg in species other than fish have also been reported . For example , Vg of the mosquito Anopheles gambiae Vg is able to interfere with the anti-plasmodium response by reducing the parasite-killing efficiency of the antiparasitic factor TEP1 [34] . Vg in honeybees also has immunological binding properties , and , moreover , is able to mediate trans-generational immunity by transporting microbe-derived molecules into developing eggs [35] . In a previous study , we have examined the interaction between L . striatellus Vg ( LsVg ) and the RSV CP and addressed the molecular interaction and its function in mediating RSV vertical transmission [5] . In the present investigation , we further explored molecular details of the RSV–Vg interaction occurring prior to ovary-uptake . We found that only the Vg produced by and processed in insect hemocytes can interact with RSV . This molecular form of Vg is also produced by non-female L . striatellus , in hemocytes only and facilitates RSV transmission .
Our previous study indicated that RSV RNPs entered the L . striatellus oocyte by binding to the Vg protein before reaching the germarium [5] . To ascertain in which tissue ( s ) the RSV–LsVg interaction occurs , we analyzed the expression levels of LsVg in various female L . striatellus tissues and carried out experiments to localize the LsVg protein . Tissues analyzed included the fat body , hemocytes , the midgut and salivary glands , all of which have been proposed previously to be involved in the transmission of persistent propagative viruses . Quantitative real-time PCR ( qRT-PCR ) was performed to test the LsVg gene expression levels . Both the fat body and hemocytes produced abundant LsVg mRNA . By contrast , only a few of the salivary gland and midgut samples exhibited LsVg expression , all at low levels ( Fig 1A ) . An immunofluorescence assay ( IFA ) was performed to visualize LsVg protein distribution . Using an LsVg-specific monoclonal antibody ( designated as Ab47Km in this study ) against the polypeptide fragment RNQQKTKSRSRRS [5] , the LsVg protein was found to localize in four types of tissue at varying abundances ( Fig 1B ) . Consistent with the observed mRNA levels , LsVg protein was abundantly synthesized in both the fat body and hemocytes ( Fig 1B ) , but was only at the detectable levels in a few of the salivary gland and midgut samples . Given that multiple L . striatellus tissues expressed LsVg and that both the fat body and hemocytes produced the protein at high levels , we next investigated whether LsVg from these tissues plays a role in mediating RSV transmission . To determine the tissue or space in which the LsVg–RSV interaction occurred , IFAs were performed to detect the co-localization of LsVg and RSV in the insect tissues . By using antibodies against the RSV RNPs , we found that the virus was distributed in all four tested L . striatellus tissues ( Fig 2 ) . To our surprise , LsVg and RSV co-localized only in hemocytes . Even though LsVg was abundant in the fat body , the protein did not co-localize with RSV in that tissue . No co-localization was observed between RSV and the low levels of LsVg in the midgut or salivary glands either . Previous studies have demonstrated that the Vg homologous protein is proteolytically cleaved into several subunits before being deposited in the eggs as vitellin ( Vn ) polypeptides [19] . Our earlier investigation of L . striatellus Vg also indicated that its N-terminal Vit-N domain does not interact with the RSV CP in vitro [5] . We thus hypothesized that LsVg is tissue-specifically processed and that the molecular form existing in the fat body is unable to interact with RSV . To test this hypothesis , we analyzed the LsVg cleavage profile and used subunit-specific anti-LsVg antibodies to investigate the subunit composition of LsVg in different tissues . Insect Vgs are usually proteolytically cleaved prior to secretion into the hemolymph , whereas the Vg subunits , in some cases , are further processed in the ovaries [17 , 19] . To reveal processing patterns of the LsVg protein in different tissues , we first analyzed the subunit composition of the L . striatellus vitellin ( LsVn ) and then used subunit-specific antibodies to determine the molecular form of LsVg in particular tissues . LsVn was purified from female insects 3 days after molting , and the LsVn subunits were fractionated by SDS-PAGE . The purified LsVn resolved into four well-separated major bands with molecular sizes of approximately 178 , 111 , 67 and 42 kDa ( Fig 3A ) . Mass spectrometry was performed to identify the LsVg-derived peptides of the four SDS-PAGE bands . Localization of the identified peptide hits revealed that the 42- and 178-kDa bands corresponded to the N- and C-parts of the LsVg protein , respectively . Peptide hits from the 67- and 111-kDa bands were included in the 178-kDa large subunit , where they clustered on its N- and C-parts , respectively . These results revealed the cleavage profile of LsVg . In particular , the precursor protein is first cleaved into small ( 42 kDa ) and large ( 178 kDa ) subunits , with the latter further cleaved into two medium-sized subunits ( 67 and 111 kDa ) . As insect Vgs are usually cleaved at the consensus tetra-residue motif RXXR by subtilisin-like endoproteases [15 , 16] , the LsVn subunit composition suggested the existence of two potential RXXR cleavage sites . Inspection of the 2 , 045 amino acids of the LsVg protein sequence allowed us to identify seven RXXR motifs: RSRR ending at amino acid 445 , RNNR at 496 , RTAR at 719 , RFQR at 844 , RNNR at 1 , 043 , RSGR at 2 , 020 and RCVR at 2 , 043 . Based on ProPeptide cleavage site prediction ( ProP 1 . 0 Server ) [36] , an above-threshold cleavage score , 0 . 873 , was obtained for the RSRR motif flanked by conserved polyserine domains ( Fig 3B ) . Cleavage after this motif resulted in two subunits with calculated molecular weights of 47 and 178 kDa . Because all the other RXXR motifs had cleavage scores below the prediction threshold , the second cleavage site was determined to be RNNR ending at amino acid 1 , 043; this inference , which was based on the molecular weight of the LsVn subunits ( 67 and 111 kDa ) as well as on the mass-spectrometry results , was further verified experimentally ( see below and Fig 3C ) . Cleavage at this motif divided the 178-kDa large subunit into two medium-sized subunits with molecular weights of 67 and 111 kDa ( Fig 3B ) . Peptide-based antibodies were produced for specific recognition of the corresponding LsVn subunits . Antibodies Ab42K and Ab111K were produced based on peptides within the mass-spectrometry-confirmed regions of the 42- and 111-kDa LsVn subunits ( Fig 3B ) . Because multiple RXXR motifs were distributed within the calculated 67-kDa LsVn subunit , two antibodies , Ab67K1 and Ab67K2 , were produced to recognize different regions of this fragment ( Fig 3B ) . Western blotting confirmed the predicted RXXR cleavage motifs . As expected , antibody Ab42K recognized the 42-kDa band , both Ab67K1 and Ab67K2 recognized the 67- as well as the 178-kDa bands , and Ab111K exhibited immunoreactive signals to both 111- and 178-kDa bands ( Fig 3C ) . The small LsVn subunit had a molecular weight ( 42 kDa ) lower than the calculated value of 47 kDa , which suggests that further processing might occur in the ovaries . Given that no additional RXXR motif was present within the 47-kDa polypeptide , how this protein processing occurs is unclear . According to the observed LsVg cleavage pattern , three subunit-specific antibodies ( Ab42K , Ab67K2 and Ab111K ) were used in the subsequent experiments to analyze the molecular forms of LsVg present in different tissues . IFA was performed using subunit-specific antibodies to detect subunit distribution . Similar to the results obtained with antibody Ab47Km ( Fig 1B ) , the use of the Ab42K antibody resulted in immunoreactive signals in both the fat body and hemocytes ( Fig 4A ) . By contrast , Ab67K2 and Ab111K produced immunoreactive signals only in hemocytes ( Fig 4A ) . These results indicated that only the N-terminus of LsVg exists in the fat body . Western blotting was performed to determine the molecular weights and the subunit composition of proteins in the fat body extract and hemolymph . Using the Ab42K antibody , a 60-kDa band was detected in both the fat body extract and the hemolymph ( Fig 4B ) . Antibodies Ab67K2 and Ab111K both recognized a strong band of approximately 200 kDa in the hemolymph ( Fig 4B ) as well as a similar but much weaker band in the fat-body extract ( Fig 4B ) . As no full-length LsVg protein was recognized by any of the three antibodies , we hypothesized that the LsVg protein , when synthesized , undergoes rapid cleavage to yield two subunits with molecular sizes of 60 and 200 kDa . While both subunits exist inside the hemocytes , only the small 60-kDa subunit exists in the fat body . The weak 200-kDa band detected by western blotting with LsVg C-terminal antibodies ( Ab67K2 and Ab111K ) in fat-body extracts may represent unconsumed proteins , or contamination from the hemolymph attached to fat-body tissues . The detected 60 and 200 kDa bands were larger than their calculated sizes ( 47 and 178 kDa ) ; this increased molecular size may be due to post-translational modifications such as glycosylation , lipidation , phosphorylation or sulfation [12 , 13] . These results also suggested that LsVg was cleaved at a single site ( RSRR ) prior to protein invasion of the ovaries . We then performed coimmunoprecipitation assay to confirm the in-vivo physical interaction between RSV and the LsVg subunits . The anti-RSV antibodies coimmunoprecipitated the LsVg large subunit , but not the small one , from the female crude extracts , confirmed the physical interaction occurred between RSV and the large LsVg subunit ( S1 Fig ) . We then performed dual immunofluorescence assay to analyze the distributional pattern of the two LsVg subunits within hemocytes . Hemocytes were probed with two different antibodies: Ab111Km antibody followed by staining with the fluorescence dye Alexa 488 , or with Ab42K antibody stained with Alexa 568 . The two types of fluorescence signals were merged in all LsVg-expressing hemocytes ( Fig 4C ) , indicating the co-existence of N- and C-terminal LsVg subunits within individual hemocytes . These results , combined with the protein sizes revealed by western blotting , indicated that LsVg exists in hemocytes as a complex of large and small subunits . To determine whether the full-length mRNA transcript or a truncated version is expressed in the fat body , we designed three pairs of PCR primers to amplify the different regions of LsVg . qRT-PCR analysis indicated that the three mRNA fragments were expressed at similar abundances ( Fig 4D ) , suggesting the production of the full-length transcript . We also used RNA interference ( RNAi ) with double-stranded RNA of the large LsVg subunit to knockdown gene expression , and then measured the expression level of the small subunit . Western blotting with antibodies Ab42K and Ab67K2 revealed dramatically decreased LsVg protein levels following gene knockdown ( Fig 4E ) , indicating a single transcript containing both the small and large subunits . We also cloned the full-length LsVg mRNA from both the hemocytes and fat body of female SBPHs . Comparison of the LsVg cDNA sequences from these tissues did not detect any variable splicing . Taken together , our results indicated that the LsVg protein is expressed in both hemocytes and the fat body of L . striatellus . After synthesis , the protein is cleaved at the RSRR motif , resulting in large and small subunits . The two-subunit complex exists in hemocytes , whereas only the small subunit remains in the fat body . How and why the large LsVg subunit is depleted in the fat body remains unclear . In addition to nourishing developing embryos , insect hemocytes and hemolymph play key roles in innate immunity . We thus investigated whether LsVg is also expressed by hemocytes of nymphal or male L . striatellus . According to a qRT-PCR analysis , both the nymphs and males expressed LsVg , and hemocytes were the only LsVg-expressing tissue ( Fig 5A ) . An IFA using the three subunit-specific antibodies ( Ab42K , Ab67K2 and Ab111K ) further confirmed the qRT-PCR results . All antibodies reacted to the corresponding subunits in hemocytes ( Fig 5B ) . A dual immunofluorescence experiment with Ab42K and Ab111Km was used to reveal co-localization of the large and small subunits ( Fig 5C ) . When expressed in hemocytes , the LsVg protein co-localized with RSV ( Fig 5B ) . To exclude the possibility that LsVg bound to various pathogens in a non-specific manner , we microinjected Escherichia coli into the L . striatellus hemolymph and looked for the localization of the LsVg protein with the phagocytosed bacteria . The bacteria did not co-localize with LsVg inside hemocytes ( Fig 5D ) , indicating that the interaction between RSV and LsVg was a specific event . Western blotting was performed to determine the subunit composition of LsVg in the male hemocytes . All the three antibodies Ab47K , Ab67K2 and Ab111K recognized a strong band of more than 200 kDa in the hemolymph ( Fig 5E ) , indicating that the LsVg in male remained in its full-length molecular form . To determine the role of LsVg in mediating RSV survival and transmission within the insects , we generated LsVg-deficient L . striatellus nymphs using RNAi with dsLsVg via microinjection . At day 6 following the onset of feeding , LsVg expression levels within hemocytes analyzed by IFA using antibody Ab111Km exhibited a dramatic reduction ( Fig 6A ) . We then assessed whether RSV transmission was affected by RNAi-mediated LsVg deficiency . The RSV titers in various L . striatellus tissues were assessed by qRT-PCR . Compared with the control group injected with dsGFP , LsVg dsRNA-treated L . striatellus showed similar RSV titers in the midgut and fat body , whereas significantly lower virus titers were observed in the hemolymph and salivary glands ( Fig 6B ) . These results indicate that LsVg functions in facilitating RSV survival in , and transmission through the hostile hemolymph environment . To confirm the function of LsVg in facilitating RSV survival in the hemolymph environment , equal volumes of LsVg or GFP dsRNA were delivered into the hemocoel of RSV-free L . striatellus nymphs via microinjection . Insects were allowed to feed on RSV-free rice seedlings . At 72 h following the microinjection , when the LsVg expression levels had been dramatically reduced by dsLsVg treatment ( Fig 6C ) , purified RSV RNPs were directly delivered into the insect hemocoel . At 1 , 6 , 12 and 24 h following virus delivery , RSV titers in the L . striatellus body were assessed by qRT-PCR . Compared with the control group treated with GFP dsRNA , LsVg dsRNA-treated L . striatellus showed significantly lower virus titers beginning 6 h post-injection ( Fig 6D ) . These results further confirmed that the hemocyte-produced LsVg plays a role in facilitating RSV survival in the hostile hemolymph environment . Immuno-blocking experiment with anti-Vg antibodies revealed similar results . Inoculation of anti-Vg antibody into hemolymoh before RSV delivery decreased RSV survival in the hemolymph , especially at the early stages ( Fig 6E , 6h and 12 h ) ; however , the difference was not significant at the later stage ( Fig 6E , 24 h ) . It might be that most of the RSVs were inside the hemocytes at the late stage , so immuno-blocking did not affect the function of the intracellular Vg .
Plant viruses have been reported to achieve vertical transmission within insect vectors via the transovarial transportation system of the insect Vg protein [5 , 37] . Traditionally , insect Vg transported into the ovaries has been thought to be synthesized in the fat body . In the current study , we demonstrated that L . striatellus hemocytes also synthesize abundant Vg protein and that only hemocyte-produced Vg interacts with RSV in vivo . By clarifying the subunit composition of LsVn and by using LsVn subunit-specific antibodies , we revealed that LsVg is synthesized and proteolytically cleaved into the N-terminal small subunit and the C-terminal large subunit in both the fat body and hemocytes . The large subunit produced in the fat body is further consumed , with the large subunit that remains in hemocytes capable of interacting with the RSV CP . Moreover , we showed that LsVg is also expressed in male and nymphal SBPHs and revealed the gender-independent function of LsVg . LsVg expressed in the hemocytes of non-female SBPHs can also interact with RSV in vivo , thus protecting the virus from the hostile hemolymph environment and facilitating its systemic infection . This study identified hemocytes , a major component of the insect immune system , as a new Vg-producing tissue . To our knowledge , this is the first study to demonstrate that an insect Vg gene is expressed by hemocytes . We also addressed the function of the hemocytes-produced Vg protein in virus recognition and transmission . The insect Vg gene has been previously reported to be expressed in tissues other than the female fat body . Although the Vg protein was not functionally investigated in most of these studies , its action has been physiologically elucidated in some cases . For example , Apis mellifera Vg is synthesized in the hypopharyngeal glands and the adjacent head fat-body cells of functionally sterile honeybee workers , implying that Vg is used for brood food production [38–40] . In Bombus hypocrita , the Vg gene is expressed at various levels in different castes , including the queen , workers and drones , from pupal to adult stages [41] . Camponotus festinatus Vg has been identified in both the queen and workers . The concentration of Vg is higher in queenless workers than in queenright ones , and Vg is synthesized at low concentrations before adult eclosion [42] . In addition to tissue-specific expression , tissue-specific processing of the Vg protein has been reported . For example , Leucophaea maderae Vg is expressed in both the male and female fat bodies . In this species , Vgs produced in the two sexes are similar in terms of their native molecular weights , but differ in their cleavage profiles and polypeptide compositions [14 , 43] . At present , little is known about the relationship between the function of Vg and its biochemical and structural properties . By addressing the functional relationship determined by the LsVg subunit/domain composition , we have confirmed the possibility that proteolytically processed Vg with different domains can mediate specific functions . To our knowledge , our investigation is one of the few studies to have classified the function of Vg according to its subunit or domain composition . The mature Vg contains an N-terminal domain ( Vitellogenin_N [Vit_N ) ] , a middle-region domain of unknown function ( DUF1943 ) , and a von Willebrand factor type D ( vWD ) C-terminal domain [5 , 33 , 44–46] . Domain Vit_N is required for interaction with the Vg receptor ( VgR ) [9 , 47] , while domains DUF1943 and vWD play roles in pathogen recognition [48] . An example in vertebrate animals is that of Oreochromis aureus Vg , which interacts with VgR via a polypeptide fragment located in the Vit_N domain [9] . Arthropoda , Macrobrachium rosenbergii Vg also interacts with its receptor via a specific β-sheet region in the Vit_N lipoprotein domain [47] . In scallops , both the recombinant DUF1943 and vWD domains of the Patinopecten yessoensis Vg protein can interact with the lipopolysaccharides and lipoteichoic acid expressed on the bacterial cell wall [45] . In insects , the L . striatellus Vg protein has been well studied by in vitro experiments that have revealed strong , weak and non-existent interactions between the RSV CP and vWD , DUF1943 and Vit_N domains , respectively [5] . In this study , we verified that the fat body-produced N-terminal Vg subunit is included in the Vit_N domain sequence . Because the L . striatellus fat body-Vg lacks any microbe-binding domains , it plays no role in mediating RSV transmission . This subunit contains both a signal peptide for secretion and a recognition site for receptor binding , however , and is thus expected to be secreted and then taken up by oocytes . Our analysis revealed the existence of an N-terminal LsVg fragment—in contrast to the complete protein—in abdominal fat body tissue . Three lines of evidence demonstrate that the N-terminal LsVg fragment is processed from the full-length LsVg protein in the fat body rather than being synthesized from a different gene . Firstly , the polyserine tracts flanking the R-X-R/K-R motif have been demonstrated to be cleaved in almost all Vg homologs ( see review [7] ) . Second , qRT-PCR analysis detected similar mRNA abundance levels between N- and C-termini of the Vg gene ( Fig 4D ) . Third , and most importantly , dsRNA of the C-terminal large subunit dramatically interfered with mRNA levels of the N-terminal small subunit ( Fig 4E ) . The significance of the existence of this Vg fragment in the fat body remains unclear; however , this type of Vg processing has also been reported in honeybees [23] . Havukainen et al . have presented a structural model for the N-terminal Vg fragment that includs a conserved β-barrel-like shape , with a lipophilic cavity and two insect-specific loops , thus indicating a capacity for lipid transport . In general , arthropod hemolymph is hostile to pathogens because it contains both circulating hemocytes and antimicrobial proteins [49–51] . Viruses acquired via the insect midgut can accordingly be transported to other tissues only after successful escape from the hostile hemolymph environment ( e . g . , through survival in plasma and evasion of hemocyte phagocytosis ) . Studies have revealed , however , that a successfully transmitted microbe can protect itself against the host immune system by exploiting host hemolymph proteins , thereby transforming the host hemolymph into a relative benign environment . Examples include Potato leafroll virus transmission by the green peach aphid Myzus persicae and Tomato yellow leaf curl virus transmission by the sweetpotato whitefly Bemisia tabaci . In these cases , the viruses gain protection and achieve successful transmission by binding to GroEL , a hemolymph protein produced by endosymbiotic bacteria of the insect [52–54] . During the transmission of West Nile Virus by Aedes aegypti , viruses can even establish infection in hemocytes by binding to the mosquito hemolymph protein that leads to the phagocytosis pathway [55] . In previous studies , Vg from fish has been revealed to have immunologic functions . By binding to the pathogen , the Vg protein can neutralize/kill the pathogen , either directly or in an indirectly fasion as opsonins mediate macrophage phagocytosis [26 , 27 , 33] . However , whether recognition by Vg of an arthropod vector would result in virus tolerance is not clear . In this study , the L . striatellus hemolymph was revealed to be a relatively benign environment for RSV . RSV tends to be cleared from the SBPH hemolymph slowly . When E . coli was delivered into the L . striatellus hemolymph through microinjection , the maximum level of phagocytosis was reached at 1 . 5–2 h post injection , and the bacterial numbers in the hemolymph ( including in hemocytes ) were dramatically decreased to <20% within 24 h ( S2 Fig ) . When RSV suspension was delivered into the hemolymph , by contrast , maximum phagocytosis was reached at about 20 h post-injection ( S2 Fig ) , and 50% of the virus was retained in the hemolymph ( including the hemocytes ) after 24 h ( Fig 6D ) . By uncovering a positive correlation between the Vg presence and the RSV in-hemolymph survival ( Fig 6 ) , this study has provided clues regarding the contribution of the RSV-LsVg interaction to the protection of RSV in the hemolymph and the development of a benign L . striatellus hemolymph environment during RSV transmission . This vector-Vg-virus interaction may be a common molecular mechanism to facilitate the passage of the virus through the vector hemolymph . In summary , we have revealed that an insect Vg protein undergoes tissue-specific processing , with the molecular form produced specifically by the hemocytes used by the virus to aid its survival in the hemolymph , thus facilitating virus transmission . Detailed analyses of the molecular interactions between the virus and its insect vector are required for the exploitation of novel virus control strategies that target specific points in the virus life cycle and interfere with virus transmission .
RSV-free and RSV-infected L . striatellus individuals used in this study were originally captured in Jiangsu Province , China , and were maintained in our laboratory . All plants used for L . striatellus rearing were grown inside a growth incubator at 25°C under a 16-h light/8-h dark photoperiod . To ensure a high offspring infection rate , viruliferous female imagoes were cultured separately , 15% of their corresponding offspring were tested for RSV infection through a dot-enzyme-linked immunosorbent assay using RSV-specific monoclonal antibodies provided by Dr Xueping Zhou ( Institute of Biotechnology , Zhejiang University [56] ) . The RSV antibody was produced using the RSV RNPs as antigen . This virion-specific monoclonal antibody was used in all the in vivo experiments that were performed to determine RSV localization or co-localization with LsVg . An insect population with a predicted infection rate of 100% was used in the experiments . For hemolymph isolation , the SBPHs were anesthetized at -20 centigrade for 3 min , and then the forelegs were severed at the coxa-trochanter joint by forceps . The hemolymph was expelled and drawn to the tip of clean forceps . Only clear droplets were collected to avoid contamination by fat body [55] . In droplets contaminated by fat body , transparent oil drops can be seen . The SBPHs were then dissected in pre-chilled PBS buffer . Insects were dissected from the abdomen , and the wound was rinsed gently two times with PBS buffer . Because the insect fat body compose of a meshwork of loose lobes suspended in the hemocoel and bathed in the insect hemolymph , it is difficult to dissect it out in its entirety and without contamination of hemolymph or other tissues . We collect most fat body without contamination from other tissues , and placed it in PBS buffer . Tissues including the midgut , salivary glands , ovaries of the female and testes of the male were washed twice in PBS to remove contaminating viruses or proteins from the hemolymph . For quantitative analysis of LsVg expression in SBPH tissues , we dissected female , male or third-instar nymphal L . striatellus and collected the tissues according to the protocol described above . RNA was extracted from the tissues of individual insects . Reverse transcriptional PCR and SYBR-Green-based qPCR were performed according to the protocols provided by the manufacturer . Primer pairs used to amplify LsVg and LsVn / LsVg subunits were LsVg-QF / LsVg-QR , 47K-QF / 47K-QR , 67K-QF / 67K-QR and 111K-QF / 111K-QR ( S1 Table ) . Viral RNA copies were measured by qRT-PCR using primer sequences pc3-F and pc3-R ( S1 Table ) , which were designed and synthesized according to the nucleocapsid protein ( Pc3 or CP ) gene sequence ( DQ333944 ) . L . striatellus elongation factor 2 ( ef2 ) was amplified as an internal control for the loading of cDNA isolated from different samples . Primers used for ef2 amplification were ef2-QF / ef2-QR ( S1 Table ) . Water was used as a negative control . Mouse anti-Vg monoclonal antibodies against Vg peptides RNQQKTKSRSRRS and RMQPLNKEEKQNVF were prepared by Abmart ( Shanghai , China ) as previously described [5] , and were designated as Ab47Km and Ab111Km in this study . To prepare LsVn subunit-specific antibodies , the LsVn subunit-specific peptides KSRRNILPQSDSNQ , AQVDSDTKHMR , YKNPGEAPELR and RMQPLNKEEKQNVF were conjugated to mcKLH and injected into rabbits , and the corresponding antiserums were prepared by GenScript ( Nanjing , China ) . The antibodies produced were designated as Ab42K , Ab67K1 , Ab67K2 and Ab111K respectively . Insects tissues were placed in PBS on silylated glass slides ( Sigma cat . no . S4651; St . Louis , MO , USA ) and allowed to dry . Tissues were then fixed in 4% paraformaldehyde at room temperature for 30 min . The slides were rinsed twice with PBS and then incubated in PBST/FBS ( PBS containing 2% Tween 20 and 2% fetal bovine serum ) for 30 min . To detect LsVg localization in different tissues , the slides were incubated with mouse anti-Vg monoclonal antibody Ab47Km ( 1:300 dilution in PBST/FBS ) for 1 h and then Alexa 568-labeled goat anti-mouse antibody ( 1:200 dilution in PBST/FBS ) for 1 h . The slides were rinsed three times with PBST , and the nucleoli were stained with TO-PRO-3 iodide ( Invitrogen cat . no . T3605; Carlsbad , CA , USA ) at room temperature for 3 min . The samples were examined using a Leica TCS SP8 confocal microscope . To detect co-localization of LsVg with RSV in the SBPH tissues , slides were prepared according to the protocol described above . The anti-RSV monoclonal antibody was labeled with Alexa Fluor 488 according to the Alexa Fluor 488 Monoclonal Antibody Labeling kit ( Invitrogen ) instructions . The slides were sequentially incubated for 1 h each with antibody Ab47Km ( 1:300 diluted in PBST/FBS ) , Alexa 568-labeled goat anti-mouse antibody ( 1:200 diluted in PBST/FBS ) and with Alexa 488-labeled anti-RSV monoclonal antibody and then stained with TO-PRO-3 iodide for 3 min . To detect co-localization of LsVn subunits with RSV , slides were prepared as described above . The slides were incubated with mouse anti-RSV and rabbit anti-LsVn antibodies ( Ab42K , Ab67K2 or Ab111K; 1:1000 dilution in PBST/FBS ) , followed by Alexa 488-labeled goat anti-mouse and Alexa 568-labeled goat anti-rabbit antibodies ( 1:200 dilution in PBST/FBS ) , and finally with TO-PRO-3 iodide for nucleolus staining . To detect co-localization of LsVg subunits in hemocytes , Ab42K and Ab111Km were used as the primary antibodies , and Alexa 568-labeled goat anti-rabbit and Alexa 488-labeled goat anti-mouse antibodies were used as the secondary antibodies . To detect co-localization of LsVg with E . coli in hemocytes , the GFP-expressing bacteria were suspended in sterile water at an OD600 of 1 . 0 . Subsequently , 13 . 8 nl of the bacterial suspension was delivered into the hemocoel of third-instar nymphs; 1 . 5 h after microinjection , the insects were dissected and the hemolymph was collected . Slides for confocal microscopy were prepared as described above . The primary antibody for LsVg detection was Ab42K and staining was performed with Alexa 568 . Newly emerged female SBPHs were allowed to grow for 3 days before extraction of Vn . Two grams of the insects were ground in liquid nitrogen into a fine powder and incubated in 1 ml of 0 . 4 M NaCl solution for 20 min at 4°C . The suspension was centrifuged at 3 , 300×g for 10 min at 4°C to remove insect debris . The supernatant was centrifuged three times at 1 , 000×g for 10 min at 4°C to remove lipid on the surface of the supernatant . The sample was then treated three times as follows: after addition of 8 ml of ddH2O , the mixture was incubated overnight at 4°C followed by centrifugation ( 1 , 000×g , 20 min , 4°C ) to precipitate the Vn protein . The precipitated protein was dissolved in 1 ml of 0 . 4 M NaCl solution and centrifuged again ( 3 , 300×g , 10 min , 4°C ) to remove any undissolved precipitate . The final supernatant was applied to a size-exclusion Superdex 200 10/300 GL column ( GE Healthcare , Piscataway , NJ , USA ) , and the fractions containing Vn of the highest concentration and purity were collected . For mass spectroscopy analysis , the purified Vn protein was separated on a 10% SDS-PAGE gel ( Bio-Rad Laboratories , Hercules , CA , USA ) and stained with Coomassie Blue ( Bio-Rad ) . The bands corresponding to the 178- , 111- , 67- and 42-kDa Vn subunits were excised and digested , and the peptides were subjected for liquid chromatography-tandem mass spectrometry analysis . Peptides were identified using an LTQ-Orbitrap XL with Easy nLC-1000 ( Thermo Fisher Scientific ) , and proteomics data were analyzed using Proteome Discoverer 1 . 4 ( Thermo Fisher Scientific ) . To confirm the subunit composition of LsVn , the purified Vn protein was fractionated on a 10% SDS-PAGE gel ( Bio-Rad ) and processed for immunoblotting . The LsVn/LsVg subunit-specific antibodies Ab42K , Ab67K1 , Ab67K2 or Ab111K ( 1:10 , 000 dilution in PBST/FBS ) was used to probe the corresponding LsVn subunits . The bound antibodies were detected by using horseradish peroxidase-conjugated goat anti-rabbit secondary antibodies ( Sigma ) , and the blots were developed using the enhanced chemiluminescence Western Blotting Detection System ( GE Healthcare ) . Western blotting was performed to measure the subunit composition and molecular sizes of LsVg in the fat body and hemocytes . Both fat body and hemolymph were dissected from the female insects ( 3 days after molting ) . Tissues from 50 insects were placed in 100 μl of PBS buffer and boiled in SDS-PAGE loading buffer . When the same amounts of total proteins were loaded to the gel , we found that LsVg was cleaved into two subunits , one small subunit of 60 kDa and one large of 200 kDa; however , the concentration of the two subunits in different tissues were very different ( S3 Fig ) . It was difficult to determine whether the large subunit existed in the fat body or whether the small subunit was secreted from the fat body . We then adjusted the western blotting loading with same amount of the LsVg small subunit as a control , and compared the amounts of the large subunit to determine subunit distribution . We performed western blotting with Ab47Km to determine the amounts of the small subunit in fat body and hemolymoh . Then fat body and hemolymph protein samples with the same amount of the small subunit were applied to the SDS-PAGE , and probed with antibodies Ab42K , Ab67K2 or Ab111K . Two DNA fragments , one specific to the coding sequences of the LsVg N-terminal small subunit and the other to the C-terminal large subunit , were PCR amplified and designated as VgN and VgC , respectively . Primer pairs used for the amplification of the VgN and VgC fragments were VgN-si-F/VgN-si-R and VgC-si-F/VgC-si-R , respectively ( S1 Table ) . DsRNA was synthesized using a commercial kit ( Ambion ) and purified by phenol:chloroform extraction and isopropanol precipitation . Finally , 36 . 8 nl of dsRNA at 1 ng/nl was delivered into the insect hemocoel for gene silencing . GFP dsRNA , which was used as a negative control , was synthesized and microinjected following the same protocol . To confirm that both LsVg subunits were expressed from the same transcript , dsRNA of VgC was delivered into the hemocoel of the fifth-instar nymphal L . striatellus individuals . The insects were cultured in new chambers with healthy rice seedlings until emergence of adults . Females were transferred to new chambers for an additional 48 h of culture . Then the insects were dissected , and the fat body protein extracts were prepared for Vg expression analysis . Western blotting was performed according to the procedure described above . The antibodies used for the detection of the fat body-expressing LsVg subunit were Ab42K and Ab67K2 . To determine the influence of LsVg on RSV horizontal transmission , dsRNA of VgC was delivered into the hemocoel of RSV-infected third-instar nymphs of L . striatellus . On day 6 of culture in new chambers , a subset of the insects was collected and the expression of Vg in their hemocytes was measured by confocal microscopy . The remaining insects were collected , and the RSV titers in various tissues were measured . Tissues , including the hemocytes , midgut , salivary glands and fat body , were collected as described above . Virus titers were determined by qRT-PCR using primer pair pc3-F / pc3-R ( S1 Table ) . To determine the effect of LsVg on RSV survival inside the hemolymph , dsRNA of VgC was delivered into the hemocoel of RSV-free third-instar nymphal L . striatellus individuals . After 72 h of culture in new chambers , some of the insects were collected , and the expression of Vg in their hemocytes was measured by confocal microscopy . Following successful knockdown of Vg expression , purified virus RNPs in PBS buffer were microinjected into the insect hemocoel . At 1 , 6 , 12 or 24 h after RSV microinjection , RNA was extracted from the whole insect body and virus titers were measured according to the protocol described above . All graphing and statistical analyses were performed using Prism 6 . 0 software ( GraphPad Software , CA , USA ) . Data were expressed as means ± standard deviation ( SD ) . The significance of differences between groups was evaluated using Student’s t-test . | Rice stripe virus ( RSV ) , which is completely dependent on Laodelphax striatellus for transmission between host plants , can also be vertically transmitted from the mother insect to its offsprings . Passing through the insect hemolymph is an essential step for both kinds of viral transmission . In this study , we found that RSV binds to L . striatellus vitellogenin ( LsVg ) in hemocytes . This tissue-specific LsVg-RSV interaction protects the virus for survival in the hemolymph and enhances both subsequent types of viral transport . More excitingly , we characterized novel and important properties of the insect form of Vg , an indispensable protein in almost all oviparous animals . In our study , we revealed for the first time that insect Vg transported into the ovary is also produced by tissues other than the fat body . Furthermore , we identified the tissue-specific molecular form of Vg responsible for its biological function in virus vertical transmission , uncovered non-female expression of Vg , and determined its function in virus horizontal transmission . These findings provide novel insights into plant-virus transmission within the vector , an important yet less explored part of the virus life cycle . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"blood",
"cells",
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"immune",
"physiology",
"immunology",
"light",
"microscopy",
"animals",
"developmental",
"biology",
"nymphs",
"microscopy",
"confocal",
"microscopy",
"antibodies",
"insect",
"vectors",
"digestive",
"system",
"research",
"and",
"analysis",
"methods",
"monoclonal",
"antibodies",
"immune",
"system",
"proteins",
"infectious",
"diseases",
"lipids",
"white",
"blood",
"cells",
"animal",
"cells",
"proteins",
"fats",
"life",
"cycles",
"exocrine",
"glands",
"insects",
"disease",
"vectors",
"hemocytes",
"arthropoda",
"biochemistry",
"eukaryota",
"cell",
"biology",
"anatomy",
"physiology",
"salivary",
"glands",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"species",
"interactions",
"organisms"
] | 2018 | Insect tissue-specific vitellogenin facilitates transmission of plant virus |
The brain constantly generates predictions about the environment to guide action . Unexpected events lead to surprise and can necessitate the modification of ongoing behavior . Surprise can occur for any sensory domain , but it is not clear how these separate surprise signals are integrated to affect motor output . By applying a trial-to-trial Bayesian surprise model to human electroencephalography data recorded during a cross-modal oddball task , we tested whether there are separate predictive models for different sensory modalities ( visual , auditory ) , or whether expectations are integrated across modalities such that surprise in one modality decreases surprise for a subsequent unexpected event in the other modality . We found that while surprise was represented in a common frontal signature across sensory modalities ( the fronto-central P3 event-related potential ) , the single-trial amplitudes of this signature more closely conformed to a model with separate surprise terms for each sensory domain . We then investigated whether surprise-related fronto-central P3 activity indexes the rapid inhibitory control of ongoing behavior after surprise , as suggested by recent theories . Confirming this prediction , the fronto-central P3 amplitude after both auditory and visual unexpected events was highly correlated with the fronto-central P3 found after stop-signals ( measured in a separate stop-signal task ) . Moreover , surprise-related and stopping-related activity loaded onto the same component in a cross-task independent components analysis . Together , these findings suggest that medial frontal cortex maintains separate predictive models for different sensory domains , but engages a common mechanism for inhibitory control of behavior regardless of the source of surprise .
Surprise occurs when expectations about the multi-sensory environment are violated . It provides an elementary cognitive and physiological process that forms the backbone of many influential theories of cognitive processing and control [1–5] . The rapid modification of ongoing actions after surprise is critical for effective goal-directed behaviors [6 , 7] . For example , while eating berries , one needs to rapidly stop ongoing actions when encountering a berry that looks , smells , or feels surprising , lest one eats a rotten berry . However , the manner in which the brain tracks surprise across different sensory domains is not fully understood . Prior imaging work has shown that unexpected events , regardless of their sensory modality , activate similar brain networks [8–11] . In line with this , scalp-electroencephalography ( EEG ) shows that unexpected events are followed by a modality-independent fronto-central P3 event-related potential [12 , ERP , 13] . The canonical neural response to surprise across modalities could indicate that the brain integrates environmental information across modalities and generates global predictions that form the basis of surprise-processing . Alternatively , surprise might result from separate , independent predictions for each sensory domain . In this latter case , the modality-independent surprise response could index a common set of downstream mechanisms triggered by surprise , regardless of sensory domain . In the current study , we tested these two alternatives against each other . While performing a cross-modal oddball task [CMO , 14] , human subjects were presented with visual or auditory unexpected events . Using the statistics of the trial sequence , we constructed two models of Bayesian surprise [5] . In one model , surprise-values were separately coded for each sensory domain ( i . e . , an unexpected sound did not reduce surprise of a subsequent unexpected visual event ) . In the alternative model , surprise was coded in a common term across modalities ( i . e . , an unexpected sound reduced surprise for a subsequent unexpected visual event ) . We fit both models to the trial-to-trial electroencephalographic response to unexpected events at each of 64 scalp-sites to determine which model better represents the neural surprise response . As mentioned above , in case this trial-to-trial modeling of the neural surprise response suggests that surprise-terms are computed separately for each sensory domain ( i . e . , surprise is not integrated into a common model ) , the expected cross-modal overlap in neural response may be explained by a common , supra-modal control mechanism that is triggered by surprise , regardless of modality . Therefore , in a second step , we aimed to test the hypothesis that the fronto-central P3 after unexpected events indexes the modality-independent activation of a cognitive control mechanism aimed at inhibiting ongoing behavior . This hypothesis was recently proposed in a theoretical framework claiming that surprise automatically engages the same motor inhibition mechanism that is recruited when ongoing actions have to be stopped [15] . The activity of this mechanism can be measured in the stop-signal task [SST , 16] , where fronto-central P3 activity following ( non-surprising ) stop-signals indexes the speed of motor inhibition [17 , 18] . To determine whether the fronto-central P3 after unexpected events in the CMO task and the P3 after stop-signals in the SST reflect the same process , we first correlated their amplitudes across tasks and subjects . We hypothesized that if they indeed reflect the same process , their amplitudes should be positively correlated . Additionally , we used independent component analysis to determine if both fronto-central waveforms load onto a common independent component [19 , 20] . In doing so , we aimed to provide converging support for the proposal that surprise-signals in frontal cortex lead to the automatic activation of a control process that aims to inhibit ongoing behavior , independent of the modality of the unexpected event .
The procedure was approved by the University of Iowa Institutional Review Board ( #201612707 ) . Fifty-five healthy young adult volunteers from the Iowa City community were recruited via a research-dedicated email list , as well as through the University of Iowa Department of Psychological and Brain Sciences’ online subject recruitment tool . The sample consisted of thirty-one females and twenty-four males ( mean age: 20 . 9 y , SEM: 0 . 05 , range 18–31 ) , eight of them left-handed . Participants were compensated with course credit or an hourly payment of $15 . Stimuli for both tasks were presented using the Psychophysics toolbox [21] ( RRID:SCR_002881 ) under MATLAB 2015b ( TheMathWorks , Natick , MA; RRID:SCR_001622 ) on an IBM-compatible computer running Fedora Linux . Visual stimuli were presented on an ASUS VG278Q low-latency flat screen monitor ( 144 Hz ) , while sounds were played at conversational volume ( ~70dB ) through speakers positioned on either side of the monitor . Responses were made using a standard QWERTY USB-keyboard . Each trial began with a central white fixation cross on black background ( 500ms ) , which was followed by an audio-visual cue ( Fig 1 ) . Participants were instructed that this cue would be informative regarding the timing of a subsequent target stimulus ( a left- or rightward white arrow ) that they would have to respond to . Participants were instructed that the cue would consist of a green circle presented in place of the fixation cross for 200ms , accompanied by a 600Hz sine wave tone of 200ms duration . There was no latency difference between the onset of the tone and the visual stimulus . After cue presentation , the fixation cross reappeared for 300ms , followed by the target ( i . e . , the target appeared exactly 500ms after cue onset ) . Participants were instructed to respond to the target as fast as possible . Target responses were collected through the keyboard ( q for leftward and p for rightward arrows ) with the index finger of the respective hand . Participants had 1 , 000ms to respond to the target , after which the fixation cross reappeared and the inter-trial interval began . The duration of the inter-trial interval lasted until 2 , 500ms from the initial onset of the fixation cross ( beginning of the trial ) was reached . Furthermore , to prevent predictable trial initiation timing , a variable-length jitter was added to the ITI ( 100 – 500ms in 100ms increments , uniform distribution ) , resulting in an overall trial duration ranging from 2 , 600ms to 3 , 000ms . After 10 practice trials without any unexpected cues , participants performed 240 trials , spread across 4 blocks . During these experimental trials , 80% of trials contained cues that were as described above ( hereafter referred to as standard cues ) . On 10% of trials , the sine-wave tone was replaced with one of 120 unique birdsong segments , which were matched in amplitude and duration to the sine-wave tone ( unexpected auditory cue; mean rise time: 3 . 56ms , SEM: 1 . 45ms ) . For these auditory unexpected cues , the visual part of the cue remained the same as for standard trials . On the remaining 10% of trials , the green circle was replaced by one of seven different geometric shapes ( upwards/downwards triangle , square , diamond , cross , hexagon , or a serifed “I”-shape ) in one of 15 different non-green colors spread across the RGB spectrum ( unexpected visual cue , cf . Fig 1 ) . This produced 105 possible unique visual cue combinations . For these visual unexpected cues , the auditory part of the cue remained the same as for standard trials . Trials were presented in pseudorandom order , with the following constraints: the three first trials of each block had to contain standard cues; no two consecutive unexpected-cue trials were allowed to occur; and each block had to have the same number of unexpected auditory and visual cues . Trials began with a white fixation cross on a gray background ( 500ms duration ) , followed by a white leftward- or rightward-pointing arrow ( go-signal ) . Participants had to respond as fast and accurately as possible to the arrow by using their left or right index finger as indicated by the direction of the arrow ( the respective response-buttons were q and p on the QWERTY keyboard ) . On 33% of trials , a stop-signal occurred ( the arrow turned from white to red ) at a delay after the go-stimulus ( stop-signal delay , SSD ) . The SSD , which was initially set to 200ms , was dynamically adjusted in 50ms increments to achieve a p ( stop ) of . 5: after successful stops , the SSD was increased; after failed stops , it was decreased . This was done independently for leftward and rightward go-stimuli: SSD started at 200ms for both left- and right-arrow trials . Then , if a stop-trial with a leftward arrow lead to a failed stop , the SSD for the next leftward arrow was decreased by 50ms , whereas the SSD for the next rightward response remained unchanged . This way , the SSD was allowed to vary independently for each arrow/response direction . Trial duration was fixed at 3000ms . Six blocks of 50 trials were performed ( 200 go , 100 stop ) . Before the main experiment , subjects practiced the task for 24 trials ( 16 go , 8 stop ) . All analysis code , as well as the task code , can be downloaded alongside the raw data at the following URL: https://osf . io/p7s32/ . For the CMO task , we quantified mean reaction time ( RT ) , mean error rate ( wrong button pressed ) , and mean miss rate ( no response made within 1 , 000ms after target onset ) for each of the three trial types ( standard cue , unexpected auditory cue , unexpected visual cue ) . We analyzed these dependent variables using a 3 x 4 repeated-measures ANOVA with the factors TRIAL TYPE ( 1–3 ) and BLOCK ( 1–4 ) . In case of a significant interaction , we performed follow-up paired-samples t-tests that compared each of the two unexpected cue conditions to the standard cue condition separately for each of the four blocks , resulting in eight total tests . The alpha-level for these comparisons was corrected using the Bonferroni correction to a corrected alpha of . 0063 ( i . e . , p = . 05 / 8 ) . For the stop-signal task , we examined the following measures: mean Go-trial RT , mean failed-stop trial RT , and mean stop-signal RT ( SSRT; computed using the integration method , Verbruggen & Logan , 2009; Boehler et al . , 2014 ) . EEG was recorded using a 62-channel electrode cap connected to two BrainVision MRplus amplifiers ( BrainProducts , Garching , Germany ) . Two additional electrodes were placed on the left canthus ( over the lateral part of the orbital bone of the left eye ) and over the part of the orbital bone directly below the left eye . The ground was placed at electrode Fz , and the reference was placed at electrode Pz . EEG was digitized at a sampling rate of 500 Hz . The CMO and SST datasets were preprocessed separately , using custom routines in MATLAB , incorporating functions from the EEGLAB toolbox [22] . The channel * time-series matrices for each task were imported into MATLAB and then filtered using symmetric two-way least-squares finite impulse response filters ( high-pass cutoff: . 3 Hz , low-pass cutoff: 30 Hz ) . Non-stereotyped artifacts were automatically removed from further analysis using segment statistics applied to consecutive one-second segments of data [joint probability and joint kurtosis , with both cutoffs set to 5 SD , cf . , 23] . After removal of non-stereotypic artifacts , the data were then re-referenced to common average and subjected to a temporal infomax ICA decomposition algorithm [24] , with extension to subgaussian sources [25] . The resulting component matrix was screened for components representing eye-movement and electrode artifacts using outlier statistics and non-dipolar components [residual variance cutoff at 15% , 26] , which were removed from the data . The remaining components ( an average of 17 . 1 per subject ) were subjected to further analyses . To investigate whether surprise is represented separately for each sensory domain ( Hypothesis 1: Cross-modal representation of surprise ) , we constructed two Bayesian surprise terms on a trial-by-trial basis , based on the trial sequences for each subject ( cf . Fig 2 ) . For both terms , the surprise value associated with an unexpected cue on a particular trial was based on the following equation [5]: Surprisei=log2 ( punexpected_cue ( 1…n ) punexpected_cue ( 1…i−1 ) ) This equation corresponds to the trial-wise Kullback-Leibler divergence between the prior probability of an unexpected cue ( denominator ) and the posterior probability of an unexpected cue ( numerator ) . In other words , the denominator is the proportion of unexpected cues on all trials prior to the current trial , while the numerator is the proportion of unexpected cues on all trials , including the current trial . For instance , if there have been 10 prior trials before the current trial , and if those trials included 1 unexpected cue , the prior probability in the denominator would be 1/10 = . 1 . After an unexpected cue on the 11th trial , the posterior probability would be 2/11 = . 1818 , resulting in a ratio of . 1818/ . 1 = 1 . 818 , and a surprise value of log2 ( 1 . 818 ) = . 8624 . Because the ratio can’t be less than 1 or greater than 2 , the Surprise parameter is bounded between 0 ( posterior = prior -> no surprise ) and 1 ( maximum surprise ) . The one exception is the first occurrence of an unexpected cue ( where the prior is zero , leading to a division by 0 inside the log ) . To address this , the surprise value for that trial was set to 1 ( maximum surprise ) . Based on this equation , we generated two different models . In Model 1 ( separate surprise terms , Fig 2 top ) , values for each sensory domain were calculated separately . In other words , the first time the subject encountered an unexpected auditory cue in the trial sequence , the surprise for that trial was 1 . Subsequent unexpected auditory cues then produced lower surprise values as the posterior and prior probabilities of unexpected auditory cues converge on the same value ( i . e . , as the ratio approaches 1 , the log approaches 0 ) with increasing numbers of previous unexpected auditory cues . Critically , these prior and posterior probabilities for auditory cues are calculated without reference to the number of prior unexpected visual cues . Thus , once a subject encounters the first unexpected visual cue , the surprise value for that trial is again 1 ( maximum surprise ) . Hence , the prior for each sensory domain is unaffected by the occurrence of unexpected cues in the other sensory domain . Note that this formulation of Model 1 assumes statistical independence in calculating these probabilities . However , the two kinds of unexpected cues were not statistically independent in the experimental design , as no trial included unexpected cues for both sensory domains . To address this , we also investigated an alternative formulation of Model 1 that respected this mutual exclusivity inherent in the experimental design . For example , upon realizing that the current trial contained an expected visual cue , this increases the prior probability for an unexpected auditory cue . It is not clear whether subjects could have reasonably learned this mutual exclusivity . Regardless of the formulation of Model 1 , the first unexpected event for either modality is maximally surprising . As a result , the model fits from this alternative formulation of Model 1 were nearly identical to the reported version , which assumed statistical independence . Therefore , we focus the report on the version that assumed statistical independence . In contrast to Model 1 , Model 2 ( common surprise term; Fig 2 , bottom ) extracted a combined surprise value , calculated without reference to sensory domain . In other words , the prior and posterior probabilities are based on the number of unexpected cues , regardless of whether those cues were visual or auditory . Note that an argument could be made that the standard-cue trials should be included in the model construction . More specifically , if the surprise term generally reflects changes in the base rates of occurrence for any event , then standard trials should be very slightly ‘surprising’ after there have been a few unexpected cue trials . However , the lack of P3 adaptation to stop-signals in the stop-signal experiment ( see Discussion ) suggests that the surprise term should only apply to events that are overtly surprising in light of task instructions , rather than any event that is novel in terms of base rates . Furthermore , including the standard trials leads to a much compromised ( i . e . , severely reduced ) fit compared to models that only consider the unexpected cues ( see S1 Fig ) . This is a result of the fact that introducing a large number of near-zero surprise trials ( i . e . , trials with standard cues ) heavily skews the distribution of the model terms towards zero . Therefore , our reported modeling results in the main manuscript only include the unexpected-cue trials . These values were then used to model the whole-brain event-related single-trial EEG response on all trials that contained unexpected cues . This was done using procedures reported by Fischer and Ullsperger [27] . For each subject , sixty-four matrices ( one for each EEG channel ) were generated that contained the event-related EEG response for each individual trial with an unexpected cue ( 24 auditory , 24 visual = 48 ) , measured in 10 consecutive time windows covering the entire cue-target interval ( 500ms , S2 Fig ) . The time windows were centered around time points ranging from 50 to 500ms and were 48ms long ( 24ms before and after the exact time point ) . EEG activity within each time window was averaged for each trial ( prior to averaging , the single-trial data were baseline-corrected by subtracting the activity ranging from 100ms– 0ms relative to the cue ) . Hence , this resulted in a matrix of 48 ( trials ) * 10 ( time points ) for each channel ( unless trials were excluded because of artifacts ) ; cf . the blue matrix in S2 Fig . Both of the two candidate surprise models constructed from the Bayesian equation were then applied to these EEG matrices ( i . e . , the average voltage data in each time window after cue-onset was modeled separately ) . In applying the models , both the surprise terms and EEG response were z-scored ( to standardize the resulting beta weights ) and the model terms were regressed onto each time-window vector of the trial by time window EEG response matrix . This was done using MATLAB’s robustfit ( ) function , which performs a linear regression that is robust to outliers . The resulting matrix of beta values was tested against 0 ( using paired-samples t-tests for the beta values , with subject as the random factor ) at each channel and time point separately . This identified channels and time periods at which the respective model surprise terms reliably captured variability in the EEG signal . This resulted in two sets of 64 ( channels ) * 10 ( time points ) = 640 individual tests ( one set for each model ) . To test which model provided a superior fit of the neural data at each channel and time-point , the resulting beta weights from each model also tested against each other , producing a third set of 640 paired-samples t-test ( again with subject as the random factor ) . To correct for multiple comparisons across these three sets of 640 t-tests , we adjusted the alpha-level using the false discovery rate correction procedure [FDR , 28] based on a family-wise alpha-level of . 01 . This resulted in an adjusted alpha-level of p = . 00044 . A detailed graphical illustration of this overall analysis strategy can be found in S2 Fig . In a separate exploratory analysis of the trial-to-trial reaction times , we similarly regressed the surprise terms from each model onto the response latencies for each target stimulus to assess whether surprise , according to each model , predicted slower responses . In addition to our above-described test of whether surprise is represented in the brain separately for each sensory domain ( Hypothesis 1 ) , we also tested whether the predicted fronto-central neural response to unexpected cues ( i . e . , the P3 ) reflects an inhibitory control signal aimed at inhibiting ongoing behavior during surprise ( Hypothesis 2 –Surprise-related frontal cortex activity reflects inhibitory control ) . To this end , we employed cross-task comparisons between the fronto-central P3 extracted for each subject from the CMO task and a separate ‘functional localizer’ task–the stop-signal task–which all subjects performed after the CMO task ( subjects performed the SST after the CMO task so they were not biased to use inhibitory control in the CMO task ) . We used two different approaches to compare activity across tasks: amplitude correlations and indepdendent component analysis ( ICA ) .
Stop-signal behavior was as expected for a sample of healthy young adults . Mean Go-RT was 520ms ( SEM: 15 . 2 ) , mean failed-stop RT was 444 . 3ms ( SEM:13 . 2 ) . Mean SSRT was 252 . 4ms ( SEM: 8 ) . Mean error and miss rates were low ( 1% and 2 . 6% , respectively ) . Mean stopping success was 51 . 4% ( SEM: . 45 , range: 46–59% ) , demonstrating the effectiveness of the adaptive stop-signal delay algorithm . In the cross-modal oddball task , correct trial RTs showed the expected pattern as well: There was a main effect of TRIAL TYPE ( F ( 2/108 ) = 25 . 3 , p = 9 . 74*10−10 , partial-eta^2 = . 32 ) , a main effect of BLOCK ( F ( 3/162 ) = 7 . 64 , p = 8 . 2567*10−5 , p-eta^2 = . 12 ) , and a significant INTERACTION ( F ( 6/324 ) = 9 . 78 , p = 6 . 51*10−10 , p-eta^2 = . 15 ) . Individual comparisons revealed that in Block 1 , both visual and auditory unexpected-cue RTs were significantly longer compared to standard-cue RTs ( t ( 54 ) = 9 . 41 , p = 5 . 48*10−13 , d = . 75 for visual and t ( 54 ) = 3 . 14 , p = . 0028 , d = . 29 for auditory , respectively ) . Furthermore , in Blocks 2 and 4 , visual unexpected-cue RTs were also longer compared to standard-cue RTs ( t ( 54 ) = 4 . 45 , p = 4 . 3*10−5 , d = . 33 and 3 . 5 , p = . 00094 , d = . 26 , respectively ) . No other comparisons survived corrections for multiple comparisons . Taken together , the data indicate the presence of an initial slowing of reaction times following unexpected cues in both modalities , which wore off over the course of the experiment ( Fig 3 ) . With regards to error rates , there was a significant main effect of TRIAL TYPE ( F ( 2/108 ) = 3 . 89 , p = . 023 , p-eta^2 = . 067 ) , with no main effect of BLOCK ( F ( 3/162 ) = . 4096 , p = . 74631 , p-eta^2 = . 0075 ) , and no INTERACTION ( F ( 6/324 ) = . 7 , p = . 65 , p-eta^2 = . 013 ) . The main effect was accounted for by lower error rates on both types of unexpected-cue trials compared to the standard-cue trials , which persisted throughout the task . With regards to miss rates , there was no significant main effect or interaction ( all p > . 14 ) .
In the current study , we tested two hypotheses about the nature of surprise processing in human frontal cortex . First , we found that fronto-central event-related activity at roughly 275-375ms following the appearance of unexpected cues tracks surprise for each sensory domain separately . Rather than incorporating surprise into a common cross-modal term , the neural response was better characterized by a model in which surprise was tracked for each domain separately . The time range and topographical extent of this activity overlaps with the well-characterized P3 trial-average ERP , which is in line with classic averaging-based ERP studies of surprise [1 , 12 , 34] . Our single-trial approach was able to disentangle two competing explanations for the common activity found for unexpected events across sensory domains , thereby providing novel insights into how frontal cortex constructs and updates models of the multi-sensory environment . We then tested whether the modality-independent fronto-central neural activity during surprise indexes a rapid inhibition of ongoing motor activity–i . e . , whether the convergence between neural signals following unexpected events , regardless of sensory domain , can be explained by a common control mechanism that is downstream from surprise . This hypothesis is relatively new [15 , 35–37] , as most previous studies of surprise focused on its cognitive effects [12 , 14 , 38 , 39] . The comparatively large sample size of our study allowed us to take the novel approach of correlating electrophysiological signal amplitudes across different tasks , revealing that the P3 amplitude following stop-signals in the stop-signal task reliably correlated with the fronto-central P3 found during multi-modal surprise . Our control analyses indicated that this correlation reflects a common process rather nuisance variables ( such as non-specific correlations of ERP amplitudes within or across tasks ) . Moreover , both ERPs reflected the same component when submitted to a joint independent components analysis . We conclude that the same process that is reflected in the stop-signal P3 is also active during cross-modal surprise . However , what is that process ? The most parsimonious explanation is that this signal reflects cognitive control within frontal cortex aimed at inhibiting ongoing motor activity . In the case of stop-signals , this stops the planned motor action , whereas in response to surprise , it produces a ‘pause’ , which purchases time for the cognitive system to update the model of the environment without continuing an action that may have been rendered inappropriate by the unexpected change in environmental demand . This pause can also be observed in the reaction time times to the subsequent target . Alternatively , the common process might reflect model updating or surprise ( as operationalized in the CMO ) . However , in the SST , stop-signals are explicitly part of the task ( and are introduced during pre-task practice ) . In other words , participants are expecting and planning for stop-signals , and their occurrence should not produce surprise . Indeed , if stop-signals were surprising , one would expect the amplitude of the stop-signal P3 to decrease as the task progressed ( i . e . , as the priors become stable and the surprise terms become smaller and smaller , which is what occurred for the fronto-central P3 in the CMO task ) . However , as the auxiliary plot in Fig 10 shows , the amplitude of the stop-signal P3 , unlike the P3 to unexpected cues in the CMO task , remained constant throughout the experiment . Therefore , surprise is unlikely to be the unifying factor that explains the commonality in brain activity across both tasks . If not surprise , perhaps the common factor is stimulus ( in ) frequency: in both the CMO task and the SST , the events in question are relatively infrequent ( 20% of trials in the CMO task include unexpected cues; 33% of trials in the SST include stop-signals ) . However , the same results that rule out surprise as the common factor rule out ( in ) frequency . Although the base rates of these ( in ) frequent events are constant in both tasks , the amplitude of fronto-central neural activity in the CMO task diminishes across the experiment , whereas it remains constant in the SST . Finally , we consider the manner in which novelty relates to surprise . There is a subtle but important distinction between surprise and novelty [40]: novelty is thought to describe a specific stimulus that is not already present in the memory system [41] , whereas surprise is a violation of prediction . Novelty and surprise often go hand-in-hand , but can be clearly dissociated . For instance , in situations with concrete priors for a specific outcome , a previously unexperienced ‘novel’ stimulus for which there is no comparison in memory can be within the realm of expectation–e . g . , an encounter of a previously unknown exotic animal at the zoo is a high-novelty , low-surprise situation . Conversely , a surprising event can be entirely non-novel . For example , finding a familiar shirt upon opening the fridge is a high-surprise , low-novelty situation . In light of this distinction , it is worth considering whether novelty might be the common factor that explains our results . Indeed , the unexpected cues in the CMO were specifically designed to be unique and novel , with the unexpected visual cue sampled from 105 unique visual cues and the unexpected auditory cue sampled from 120 unique auditory cues . In contrast , the ( in ) frequent stop-signals were anything but novel , being sampled from just two possibilities: a leftward pointing red arrow or a rightward pointing red arrow . For the SST , these two possibilities were demonstrated prior to the experiment , and participants experienced dozens of instances of the two stop-signals in the first half of the SST . Nevertheless , for the second half of the experiment , stop-signals consistently produce a large positive deflection ( cf . Fig 10 ) . Thus , novelty does not appear to be the common factor . Our preferred interpretation of the common process in terms of motor control is supported by recent studies , which found that unexpected perceptual events lead to a broad , reactive suppression of the motor system , as measured using transcranial magnetic stimulation [37 , 42] . Additionally , measurements of isometrically exerted force have shown that unexpected events lead to a rapid , reactive reduction of such steadily exerted motor activity [36] . Furthermore , unexpected events have been found to interrupt ongoing finger-tapping [43] . Finally , studies using optogenetics have shown that when regions of the subcortical network that cause inhibition of motor activity are experimentally inactivated , unexpected events no longer yield interruptive effects on motor behavior [44] . All these studies show that surprise , in addition to its prominent cognitive effects , also lead to interruption of ongoing motor activity . The interpretation that the common process between the stop-signal and CMO tasks is motor control is also supported by some features of our data . Specifically , our behavioral data indicated an incidental slowing of reaction times to the target in the CMO task when that target was preceded by unexpected cues , which is in line with prior behavioral studies [45–47] . Our exploratory analysis showed that during the task period in which this RT effect was present , the surprise model ( specifically , the separate-term model that also provided the best fit to the neural data ) was positively related to the RT data: trials with more surprising cues , according to the Bayesian model , yielded longer reaction times to the subsequent target . We propose that this extra time reflects a momentary suppression of the motor system produced by the unexpected event . Supporting this claim that this ‘pause’ is an adaptive process , accuracy was also increased following unexpected cues ( i . e . , a speed-accuracy tradeoff was enacted after unexpected cues , which may be enabled by the transient pause in the motor system that we purport to be reflected in the fronto-central P3 ) . In that vein , one notable observation is that while the surprise term fit the neural data for both domains to similar degrees ( Fig 6 ) , the trial-average response to unexpected auditory cues in our current study appeared to be larger in amplitude compared to unexpected visual cues ( Fig 5 ) . Interestingly , the reverse was the case in the reaction time pattern , where visual unexpected cues seemed to have larger effects ( Fig 3 ) . While we are hesitant to make strong conclusions based on the trial-average data , it is notable that the timing of the P3 to the different stimuli also differs in latency , which likely reflects the fact that early auditory processing is faster than visual processing [48] . Since the increase in trial-averaged P3 response to unexpected visual cues extends to a time period much closer to target presentation ( compared to the P3 to auditory unexpected cues , cf . Fig 5 ) , it is tempting to assume that this may explain the difference in RT effects . However , further studies are necessary to explicitly test this hypothesis . Finally , it is worth mentioning that motor inhibition is not the only process that is triggered by unexpected events: such events are known to trigger a cascade of ( inhibitory and excitatory ) processes , including a reorienting of attention [3 , 49] , autonomic arousal [50] , a shift towards exploratory behaviors [51 , 52] , a re-evaluation of learned associations [53] and a potential interruption of ongoing cognitive processes [54 , 55] . From the current study , as well as other past studies [36 , 42 , 43] , it is now becoming increasingly obvious that motor inhibition is a part of this multi-faceted cascade of processing , which–in concert–enables the cognitive system to flexibly react to unexpected changes in the environment . There is some debate in the literature about the interpretation of the surprise term used in our model comparison analysis . We followed the nomenclature of Itti and Baldi [5] , who termed the calculation of the Kullback-Leibler divergence of the posterior and prior probability distributions ( Equation 1 ) ‘Bayesian surprise’ . However , other authors have interpreted this term as ‘model updating’ , rather than surprise [56] . Instead of KL divergence , they favor Shannon-based information theoretical quantifications of surprise [i . e . , surprise is quantified as the inverse of the log-scaled prior expectation of a given stimulus , 57] . In past EEG studies , Shannon-based surprise has been associated with the amplitude of the centro-parietal P3 ERP [58 , 59] , rather than the fronto-central P3 examined here . This is in line with BOLD activation of parietal cortex , which tracks such Shannon-surprise in fMRI [56] . Conversely , trial-by-trial indices of Bayesian surprise are associated with the fronto-central P3 [59] , which is in line with the current study , as well as with fMRI work showing that BOLD activity in medial frontal cortex tracks Bayesian surprise [56] . Collectively , these results underscore that Shannon-surprise and Bayesian surprise are not only different computational terms but that they may be related to different neural signals . However , in terms of the theoretical distinction between Bayesian surprise and Shannon surprise , it is important to note that both concepts are closely related in most circumstances–i . e . , whenever there is surprise , it will lead to the updating of internal models of the environment . This is also reflected in a high correlation between Shannon- and Bayesian surprise that is present in most experimental circumstances ( including the current one ) . Under some circumstances , it is possible to untangle surprise and model updating by introducing different degrees of volatility into the environment [60] or by explicitly instructing participants that certain surprising cues should not be used to update the internal model of the task [56] . However , in studies like the current one , the two terms are largely identical , with the exception being trials in which in an unexpected cue follows a prolonged sequence of expected cues . ( Such trials introduce non-monotonous upticks in the Shannon surprise term , whereas the Bayesian surprise / model updating term is always monotonically decreasing ) . Perhaps most relevant is the question which term better reflects the commonplace meaning of ‘surprise’ in the everyday world , outside of the laboratory , and which term better reflects the participants’ approach to the experiment . If subjects place strong emphasis on the recent trial sequence and dynamically adapt to the changing local probabilities of unexpected cues , then the Shannon term may provide a better characterization of surprise . This would be the case if participants assume that the current environment constantly changes ( i . e . , high volatility ) . However , if subjects approach the experimental task as a specific , unchanging environment that they need to adapt to by learning the base rates of occurrence , then the Bayesian surprise term may provide a better characterization of surprise . In the current study we assumed that the latter is the case ( indeed , the experimental design involved a stable procedure for each task ) , and as such , ‘surprise’ and ‘model updating’ are essentially synonymous in our study . Taken together , our study suggests that when an environmental model is updated because of an unexpected cue , this leads to surprise , which is accompanied by inhibitory control of the motor system . From a real-world perspective , it makes sense for the cognitive apparatus to operate this way . Because we interact with the environment by executing motor commands , it is important that we interrupt ongoing motor behavior while the model of the environment is updated; ongoing actions need to be re-evaluated in light of changing environmental contingencies . We hypothesize that motor inhibition prevents the execution of actions that were appropriate under the old , now outdated model , and may also free up resources to rapidly initiate appropriate new actions . This interpretation of the medial frontal cortex is in line with prior findings regarding its role in the control of behavior [2 , 61 , 62] . Here , we propose a specific neural mechanism by which such control of behavior is achieved during surprise . In conclusion , we found that surprise-based model updating in frontal cortex occurs separately for each sensory domain , but shares a supra-model control mechanism that likely involves the inhibitory control of behavior . These results suggest a specific control mechanism that is rapidly deployed when the model of the environment unexpectedly changes . | Surprise is an elementary cognitive computation that the brain performs to guide behavior . We investigated how the brain tracks surprise across different senses: Do unexpected sounds make subsequent unexpected visual stimuli less surprising ? Or does the brain maintain separate expectations of environmental regularities for different senses ? We found that the latter is the case . However , even though surprise was separately tracked for auditory and visual events , it elicited a common signature over frontal cortex in both sensory domains . Importantly , we observed the same neural signature when actions had to be stopped after non-surprising stop-signals in a motor inhibition task . This suggests that this signature reflects a rapid interruption of ongoing behavior when our surroundings do not conform to our expectations . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"reaction",
"time",
"brain",
"electrophysiology",
"social",
"sciences",
"electrophysiology",
"brain",
"neuroscience",
"cognitive",
"neuroscience",
"clinical",
"medicine",
"cognition",
"brain",
"mapping",
"bioassays",
"and",
"physiological",
"analysis",
"vision",
"event-related",
"potentials",
"neuroimaging",
"electroencephalography",
"research",
"and",
"analysis",
"methods",
"frontal",
"lobe",
"imaging",
"techniques",
"cerebral",
"cortex",
"clinical",
"neurophysiology",
"behavior",
"electrophysiological",
"techniques",
"psychology",
"sensory",
"cues",
"anatomy",
"physiology",
"biology",
"and",
"life",
"sciences",
"sensory",
"perception",
"cognitive",
"science",
"neurophysiology"
] | 2019 | Frontal cortex tracks surprise separately for different sensory modalities but engages a common inhibitory control mechanism |
Legionella pneumophila is an environmental bacterium and the causative agent of Legionnaires’ disease . Previous genomic studies have shown that recombination accounts for a high proportion ( >96% ) of diversity within several major disease-associated sequence types ( STs ) of L . pneumophila . This suggests that recombination represents a potentially important force shaping adaptation and virulence . Despite this , little is known about the biological effects of recombination in L . pneumophila , particularly with regards to homologous recombination ( whereby genes are replaced with alternative allelic variants ) . Using newly available population genomic data , we have disentangled events arising from homologous and non-homologous recombination in six major disease-associated STs of L . pneumophila ( subsp . pneumophila ) , and subsequently performed a detailed characterisation of the dynamics and impact of homologous recombination . We identified genomic “hotspots” of homologous recombination that include regions containing outer membrane proteins , the lipopolysaccharide ( LPS ) region and Dot/Icm effectors , which provide interesting clues to the selection pressures faced by L . pneumophila . Inference of the origin of the recombined regions showed that isolates have most frequently imported DNA from isolates belonging to their own clade , but also occasionally from other major clades of the same subspecies . This supports the hypothesis that the possibility for horizontal exchange of new adaptations between major clades of the subspecies may have been a critical factor in the recent emergence of several clinically important STs from diverse genomic backgrounds . However , acquisition of recombined regions from another subspecies , L . pneumophila subsp . fraseri , was rarely observed , suggesting the existence of a recombination barrier and/or the possibility of ongoing speciation between the two subspecies . Finally , we suggest that multi-fragment recombination may occur in L . pneumophila , whereby multiple non-contiguous segments that originate from the same molecule of donor DNA are imported into a recipient genome during a single episode of recombination .
While all bacteria reproduce clonally , some also import DNA from other organisms into their chromosomes through processes such as recombination or horizontal gene transfer . The imported DNA can either replace a homologous segment of the genome ( homologous recombination ) or comprise novel genes that are new to the recipient genome ( non-homologous recombination ) . The former results in the replacement of genes with alternative allelic variants and requires the DNA to be highly similar , and possibly identical , at both ends of the fragment [1] . For this reason , homologous recombination usually occurs between closely related bacteria . The importance of recombination in bacterial evolution first became clear through the analysis of multi-locus sequence typing ( MLST ) data , which showed that phylogenetic trees constructed from individual MLST genes were often incongruent [2] . These analyses also predicted that the rate of homologous recombination varies considerably between different species [3] . There are a number of hypotheses regarding why bacteria engage in homologous recombination [4] . One explanation is that recombination is used as a mechanism by which DNA damage can be repaired using foreign DNA as a template [5] . Another is that it is a side effect of DNA uptake for use as an energy source or for DNA synthesis from nucleotide precursors [6] . Third , the ability of recombination events to remove deleterious mutations and rapidly introduce combinations of advantageous mutations could mean it increases the efficiency of natural selection and is selectively maintained [7] . Finally , a recent study has also suggested that bacteria use recombination to delete selfish mobile genetic elements from their genomes [8] . In recent years , the availability of whole genome sequence ( WGS ) data from multiple closely related bacterial isolates has enabled homologous recombination to be studied in great detail in species such as Streptococcus pneumoniae [9 , 10] , Chlamydia trachomatis [11] and Neisseria meningitidis [12] . These studies have confirmed that homologous recombination plays an important role in the evolution and adaptation of important bacterial pathogens , for example by facilitating vaccine escape [9] and antibiotic resistance [10] in S . pneumoniae . Legionella pneumophila is an environmental bacterium that parasitizes and replicates inside protozoa in freshwater and soil habitats [13] . It also now colonises man-made water systems from which humans can become infected via inhalation of contaminated aerosols [14] . Infection can cause Legionnaires’ disease , a serious and potentially fatal pneumonia [15] . L . pneumophila was first reported to have a clonal population structure based on multi-locus enzyme electrophoresis ( MLEE ) analysis [16] . However , the three primary mechanisms of bacterial recombination ( conjugation , transduction and transformation ) have since all been described in L . pneumophila [17–19] , and thus it was unsurprising when later studies reported its occurrence . Indeed , an early genomic study of the first sequenced genomes of L . pneumophila showed that recombination events are frequent and suggested that it can involve large chromosomal fragments of over 200kb [20] . More recently , larger genomic studies have demonstrated that >95% of single nucleotide polymorphisms ( SNPs ) detected within some lineages have been imported via recombination [21 , 22] . The occurrence of recombination within L . pneumophila populations has also led to the existence of multiple genetic subtypes within single outbreaks [21 , 23] . However , despite its major role in L . pneumophila evolution , the relative frequency and biological effects of recombination , such as its impact on virulence or adaptation of L . pneumophila to new niches , remain poorly understood . Here , we disentangled events arising from homologous and non-homologous recombination in six major disease-associated sequence types ( STs ) of L . pneumophila , and subsequently performed a detailed characterisation of the dynamics and biological impact of homologous recombination on L . pneumophila evolution . Our findings provide novel insights into the selection pressures of L . pneumophila and the dynamics of genomic flux within the species .
To investigate the relative contribution of homologous recombination to diversity in each of six major disease-associated lineages of L . pneumophila ( STs 1 , 23 , 37 , 42 , 62 and 578 ) , sequence reads from isolates ( n = 291 ) ( S1 Table ) were first mapped to a reference genome of the same ST ( Table 1 ) . Isolates belonging to STs that have previously been shown to be derived from the ST1 lineage were also included with ST1 isolates [22] . Gubbins was used to detect recombined regions in each of the six genome alignments [24] . This tool uses increased SNP density on branches of a phylogenetic tree as a marker , and is well suited to these six lineages that contain low background diversity . However , it should be noted that recombination between highly similar isolates may be missed , potentially leading to an underestimation of the recombination rate . In our previous study using four of these alignments with the same , or largely the same , isolates [22] , this programme showed high concordance with another recombination detection tool , BRATNextGen [25] . Detection of recombined regions using BRATNextGen is based on sequence similarity rather than SNP density , and thus the high concordance between the two different approaches provides confidence in our predicted regions . As previously reported [21 , 22] , over 96% of SNPs in STs 1 , 23 , 37 , 62 and 578 were predicted to be derived from recombination events . Furthermore , 99% of SNPs in the ST42 lineage were found in recombined regions in the present study . Thus , in all six lineages the proportion of SNPs derived from recombination is higher than that reported for the highly recombinogenic S . pneumoniae PMEN1 lineage ( 88% ) [9] and between N . meningitidis ST60 strains ( 94 . 25% ) [12] . The number of vertically inherited SNPs that remained after the removal of recombined regions in each of the six L . pneumophila lineages ranged from 94 ( ST42 ) to 1 , 006 ( ST1 ) ( Table 1 ) . Any regions detected by Gubbins that overlapped with either predicted mobile genetic elements ( MGEs ) or repeat regions ( S2 Table ) were subsequently excluded , in order to determine the sole contribution of homologous recombination to L . pneumophila diversity ( Table 2 ) . We found that between 33% ( ST62 ) and 80% ( ST578 ) of all SNPs were predicted to be in regions derived from homologous recombination events ( Fig 1A ) . However , the average length of each individual genome affected by this process varied between just 1 . 2% ( ST42/578 ) and 3 . 9% ( ST1 ) ( Table 2 ) . It should be noted that the number of SNPs predicted to be from homologous recombination might be slightly over-estimated ( and the number of de novo mutations slightly under-estimated ) since de novo mutations may have occurred on top of recombined regions . However , the error should be no more than 1 . 2–3 . 9% , in proportion with the average length of genome affected by homologous recombination events . Furthermore , detectability of homologous recombination events could also be affected by lineage diversity ( i . e . events may be more difficult to detect on longer tree branches where background SNP density is higher ) . However , because the number of SNPs associated with recombining regions is much higher compared with the background vertically inherited SNPs in all lineages , we think that any effect will be minimal . In each of the six lineages , the relative number of homologous recombination events to vertically inherited mutations ( ρ/θ ratio ) was calculated per branch for each phylogenetic tree ( S1 Fig ) and also for each lineage as a whole . The overall ρ/θ ratio for each lineage ranged from 0 . 03 ( ST37 ) to 0 . 27 ( ST23 ) , indicating that recombination events have occurred less frequently than vertically inherited mutations in all six lineages , despite bringing in between 20 . 8 ( ST37 ) and 93 . 8 ( ST23 ) times as many SNPs ( Table 2 ) . A similar ρ/θ ratio of 0 . 124 was reported in previous analysis of 25 diverse L . pneumophila genomes as inferred by an alternative recombination detection algorithm , ClonalFrame [29] . The distribution of per-branch ρ/θ ratios also differ significantly between lineages ( Kruskal-Wallis test , p<0 . 05 ) , highlighting different rates of recombination in the six major disease-associated STs . These differences could indicate variation in the biological niches of these different lineages , about which very little is currently understood , and/or the availability of recombination opportunities . To determine the relative impact of vertically inherited mutations and homologous recombination events on the coding sequence , the types of changes caused by the two processes were analysed ( Fig 1B ) . Vertically inherited mutations have resulted in approximately twice as many non-synonymous SNPs than synonymous SNPs , a result that is expected by chance when mutations occur at random in the genome and before selection has time to act on all but the most deleterious mutations . Interestingly though , the results are reversed for homologous recombination events , which resulted mostly in synonymous mutations . However , this observation is not unexpected given that variants in sequences that are horizontally transferred between different lineages will have been subjected to a longer period of evolution and selection , which has purged harmful , non-synonymous mutations . The same phenomenon has also been observed in a previous study by Castillo-Ramirez et al . ( 2011 ) [30] . Furthermore , fewer SNPs that result in a stop codon were brought in by homologous recombination events than by vertically inherited mutations , which can also be explained by this process . The lengths of the recombined regions have an approximately exponential distribution ( rate of decay = 7 . 52 x 10−5 bp-1 ) , with the majority of events being small ( <10 , 000bp ) and large events occurring relatively infrequently ( Fig 1C ) . The median recombination fragment length in each of the six lineages ranged from 5 , 613bp ( ST578 ) to 12 , 757bp ( ST37 ) , while the largest predicted region is 94 , 790bp ( ST37 ) . Large recombination segments have also been found in other species , such as Clostridium difficile [31] , Streptococcus agalactiae [32] and Streptococcus pneumoniae [33] . In the latter , a similar distribution of fragment sizes as the one described here for L . pneumophila was also reported , suggesting that transformation is optimised for exchanging short sequences rather than large features such as complete operons [33] . This scenario could be favoured as it allows for larger numbers of potentially advantageous allele combinations to be tested . Next , we determined whether there are any genomic regions that are associated with a higher number of homologous recombination events , which could reveal genes that are under diversifying selection pressure . We thus calculated the number of events predicted by Gubbins that overlap with each gene with respect to the reference genomes of the six disease-associated STs . A total of 32 hotspot regions were defined ( see Materials & Methods ) , including at least one in all six disease-associated STs ( S3 Table ) . A total of 10 hotspot regions were identified in the ST1 lineage and , remarkably , one region contained genes that are predicted to have been involved in up to 27 recombination events ( Fig 2A ) . By contrast , in the other five STs , the highest number of events affecting genes ranged from 2 ( ST37/ST578 ) to 4 ( ST42/ST62 ) . We acknowledge that the number of recombination events detected per gene , particularly in hotspot regions , could be slightly underestimated due to the possible occurrence of overlapping or nested recombined regions imported on the same branch of the phylogenetic tree . Gubbins is likely to predict these as single rather than multiple events , and genomic regions with a higher number of recombination events could be disproportionately affected . Nevertheless , the identification of hotspot regions provides good evidence that the effect of recombination in L . pneumophila is to increase the genetic diversity available for natural selection to work on , and that this diversifying selection acts non-randomly on the genome . To predict the origin of the homologous recombination regions , 536 L . pneumophila genomes were first divided into clusters using hierBAPS [47] , which were mapped onto a phylogenetic tree ( Fig 3 ) . The genomes comprise those belonging to isolates from the six major disease-associated STs ( n = 291 ) ( S1 Table ) and others from a large , species-wide collection ( n = 245 ) ( S5 Table ) . Eight BAPS clusters were identified , seven of which comprised isolates from the L . pneumophila pneumophila subspecies ( BAPS clusters 1–6 , 8 ) , and one with isolates from L . pneumophila fraseri ( BAPS cluster 7 ) . Of the 318 homologous recombination events greater than 500bp predicted in the six major disease-associated lineages , potential donors were predicted for 292 ( 91 . 8% ) ( see Materials & Methods ) . Many of the hits were almost perfect matches with 122 ( 41 . 8% ) of the fragments having over 99 . 9% nucleotide identity , and 155 ( 53 . 1% ) having hits that covered the full length of the recombination fragment ( S6 Fig ) . The number of homologous recombination events in each of the six major disease-associated lineages that were predicted to be derived from each of the eight BAPS clusters were calculated and visualised in a heat plot ( Fig 4A ) . Any events with equally good hits ( i . e . with the same nucleotide similarity and fragment length covered ) to isolates in more than one BAPS cluster were discarded for this analysis ( “No donor assigned” ) . The heat plot illustrates that , in five of the six STs , recombination donors most often belonged to the same BAPS cluster as the recipient . This is an expected finding since homologous recombination requires high , or even perfect , sequence similarity between the donor and recipient at both ends of the recombination fragment [1] , a scenario which is more likely between closely-related bacteria . The exception is ST37 in which the highest number of recombination fragments is derived from BAPS cluster 4 , although its own cluster ( BAPS cluster 3 ) accounted for the second highest number . However , all STs , with the exception of ST578 , are also predicted to have acquired recombination fragments from clusters other than their own , demonstrating the occurrence of homologous recombination between major clusters of the L . pneumophila pneumophila subspecies . This result is confirmed by the fastGEAR analysis of the prominent ST1 hotspot region , which demonstrates the sharing of alleles between different BAPS clusters ( S4 Fig ) . Overall , the finding suggests that different clades have at least partially shared the same ecological niche and perhaps even the same individual host cells in which recombination may have occurred . Importantly , this freedom of genomic exchange has provided potential opportunities for new adaptations to be shared freely amongst different clusters , which we hypothesise has been an important factor in the recent emergence of multiple major disease-associated STs from diverse genomic backgrounds [22] . Interestingly , some BAPS clusters act frequently as donors to other clusters ( e . g . BAPS clusters 4 and 5 ) , while others hardly donate except to isolates of their own cluster ( e . g . BAPS clusters 2 and 3 ) ( Fig 4A ) . Similar patterns whereby different lineages donate and receive DNA at different rates have also been observed in other species such as S . pneumoniae [10] , C . trachomatis [11] and E . coli [48] . Furthermore , just two events ( one each in ST23 and ST62 ) are derived from the L . pneumophila fraseri subspecies ( BAPS cluster 7 ) . Given that this lineage shares less than 95% nucleotide identity with the L . p . pneumophila subspecies , this was not an unexpected finding , given the high level of similarity required for homologous recombination . It could be that these two subspecies have gradually diverged due to differing ecologies , and that eventually they may become different species that are fully incapable of exchange via homologous recombination . For all homologous recombination events detected in the six STs , the nucleotide identity between the imported fragment and the recipient genome that was replaced by the fragment was calculated ( Fig 4B ) . This was to investigate the divergence levels between recombining bacteria , but it also provided a means of verifying our predictions of the recombination donors . This analysis showed that 70% of homologous recombination events occurred between closely related isolates with >98% nucleotide similarity in the affected region , which agrees with our previous finding that most fragments are derived from the same BAPS cluster as the recipient . Interestingly , two peaks can be observed at ~98% identity and ~99 . 5–100% identity . These levels of divergence correspond to the nucleotide similarity observed between isolates belonging to different clusters or the same cluster , respectively ( Fig 4C ) , and thus they represent recombination between and within clusters . It is also interesting to note that the distribution of pairwise nucleotide similarities of genomes from different clusters has a major peak around ~98% ( Fig 4C ) , which aligns with previous findings that homologous recombination tends make clusters equidistant from each other [49 , 50] . Recombination hotspot regions were next re-analysed to investigate whether the hotspots were driven by recombination events from the same or different BAPS clusters . The analysis focused on the ST1 lineage , which was previously found to contain the highest number of recombination events and the most prominent hotspots . The most notable hotspot region ( hotspot 6 ) , which was found to contain genes involved in up to 27 recombination events , was found to be driven mostly by recombination regions derived from the same BAPS cluster to which ST1 belongs ( BAPS cluster 2 ) ( S7 Fig ) . However , a small number of recombination events that are predicted to be from BAPS cluster 5 were also observed in this region . While the analysis of this region using fastGEAR is not directly comparable to the results using Gubbins , it does also suggest that the recombined regions have been imported from both the same and different BAPS clusters ( S4 Fig ) . Meanwhile , while some of the recombination events affecting the LPS locus ( hotspot 3 ) could not be assigned a donor , others were derived from BAPS clusters 1 , 2 and 5 , suggesting that high diversity in this region may be especially important . Hotspot 4 appears to be driven by recombination events from BAPS clusters 5 , 6 and 8 and contains no events derived from BAPS cluster 2 ( to which ST1 belongs ) . However , the small number of events with predicted donors in most of these hotspots limits the conclusions that can be made . Finally , the homologous recombination events that were predicted within the ST1 lineage were mapped onto the phylogenetic tree together with information regarding their predicted origin ( Fig 5 ) . This was to search for evidence of multi-fragment recombination , a process in which multiple non-contiguous segments that originate from the same molecule of DNA are imported into a recipient genome in a single episode of recombination . This process is well documented in S . pneumoniae [33 , 51 , 52] . Since the recombining fragments are non-contiguous , Gubbins will detect these as separate events although the events should be predicted to have occurred on the same branch and have the same predicted donor . Indeed , we found some evidence for the occurrence of this process in L . pneumophila , since many events with the same predicted donor , down to the BAPS cluster level and even the individual isolate level , are co-localised on branches ( Fig 5 ) . For example , 8 recombinant regions distributed throughout the chromosome that occurred on the terminal branch leading to ST1_28 are predicted to have originated from BAPS cluster 4 , and more specifically , a strain ( or multiple strains ) closely related to EUL 25 ( ST44 ) ( S8 Fig ) . Furthermore , some of these imported regions also share very similar SNP densities with respect to EUL 25 ( i . e . 5 events have SNP densities from 0–0 . 06% and 3 events have SNP densities from 0 . 28–0 . 33% ) , reinforcing the possibility that some of these recombining fragments could have been acquired from the same donor in the same event . However , it could also be that the recombining isolates have shared a common niche for a prolonged period of time , and that multiple independent recombination events have occurred during this time . Thus , further experimental studies will be required to confirm the occurrence of this process in L . pneumophila . In summary , this study has demonstrated a major role for homologous recombination in shaping the population structure and evolution of L . pneumophila , and provided detailed insights into recombination dynamics within the species . We predict that homologous recombination has played a critical role in the emergence of this environmental bacterium as an important human pathogen and suggest that future studies are required to further delineate the role of homologous recombination in the virulence and adaptation of L . pneumophila to modern , man-made environments .
L . pneumophila isolates belonging to six major disease-associated lineages are primarily used in this study ( n = 291 ) , all of which have been previously sequenced [21 , 22 , 27 , 53–55] . These include 81 ST1 ( or ST1-derived ) , 42 ST23 , 72 ST37 , 15 ST42 , 35 ST62 , and 46 ST578 isolates ( S1 Table ) . A further 245 L . pneumophila isolates , which belong to a range of STs , were also used in the inference of recombination donors ( S5 Table ) . WGS data from all but five of these isolates have been published [20–22 , 26–28 , 55–60] . Importantly , these include a set of genomes that were selected for sequencing using sequence-based typing ( SBT ) data , analogous to MLST , with the aim of encompassing as much of the species diversity as possible [55] . Accession numbers or references for all genomes are provided in S1 Table and S5 Table . Sequence reads from isolates belonging to each of the six disease-associated STs ( 1 , 23 , 37 , 42 , 62 and 578 ) were mapped to a reference genome of the same ST to enable each lineage to be studied at a high resolution . The complete genomes of Paris [26] and Alcoy [28] were used for ST1 and ST578 , and reference genomes previously generated using a Pacific Biosciences ( PacBio ) RSII sequencer were used for STs 23 ( EUL 28 ) , 37 ( EUL 165 ) , 42 ( EUL 120 ) and 62 ( H044120014 ) [27] . All six reference genomes were annotated using an in-house pipeline at the Wellcome Trust Sanger Institute ( WTSI ) , which uses Prokka [61] . The four annotated reference genomes obtained using PacBio sequencing are available from the European Nucleotide Archive under the accession numbers GCA_900119755 . 1 ( EUL 28 ) , GCA_900119775 . 1 ( EUL 165 ) , GCA_900119785 . 1 ( EUL 120 ) and GCA_900119765 . 1 ( H044120014 ) . Repetitive regions over 100bp were detected in the six reference genomes using repeat-match from MUMmer v3 . 0 [62] ( S2 Table ) . All processing and sequencing of genomic DNA from the five newly sequenced isolates was performed by the core sequencing facility at the WTSI . Paired end libraries were created as described previously [63] and samples were sequenced using the Illumina HiSeq platform and paired-end reads of 100 bases . Sequence reads of all isolates belonging to the six major disease-associated STs under study were mapped to the appropriate reference genome of the same ST using SMALT v0 . 7 . 4 ( available at: http://www . sanger . ac . uk/science/tools/smalt-0 ) . All isolates used in the study ( n = 536 ) were also mapped to the Paris ( ST1 ) reference genome [26] in order to study the species-wide phylogenetic structure . An in-house pipeline at the WTSI was used to call bases and identify SNPs as previously described [64] . All assemblies were produced from the Illumina data using a pipeline developed by the Pathogen Informatics team at the WTSI . This firstly uses Velvet Optimiser ( https://github . com/tseemann/VelvetOptimiser ) to determine the optimal kmer size before using Velvet to produce the assembly [65] . The assembly was further improved using SSPACE [66] to scaffold the contigs of the assembly and GapFiller [67] to close gaps of 1 or more nucleotides . Recombined regions were detected in the alignments of the six disease-associated STs using Gubbins [24] . Phylogenetic trees of these lineages were generated using RAxML v7 . 0 . 4 [68] , firstly using all SNPs to later allow ancestral sequence reconstruction , and secondly using only the vertically inherited SNPs ( i . e . excluding SNPs in recombined regions ) . A phylogenetic tree of the total 536 isolates was constructed using all the detected SNPs , as the high diversity of the whole collection renders recombination detection very difficult . In all cases , the GTR+GAMMA method for among site rate variation was used and 100 bootstrap replicates were performed to assess support for nodes . The alignment of all 536 genomes against the Paris reference genome was also used to group the isolates into clusters using hierBAPS [47] . The annotation files from each of the six reference genomes were parsed to detect regions annotated as “integrase” , “transposase” , “recombinase” , “phage” , “lvrA” , “csrA” , “HTX” , “helix-turn-helix” , “xre” , “conjugal” , “conjugation” , “tra” , “trb” , “vir” and “mobile” . Both the published annotation files of the Paris ( ST1 ) and Alcoy ( ST578 ) complete genomes and those generated using the in-house pipeline at the WTSI were used . However , the new annotations were only considered when the original one was a “hypothetical protein” in order to respect experimentally proven annotations . Plots showing the mapping coverage of each isolate against the corresponding reference genome were also evaluated . Regions over 8kb with no coverage and that did not match repetitive regions were considered as potential mobile regions . Other software to detect mobile genetic elements ( MGEs ) was also used including AlienHunter [69] and Island Viewer , the latter of which incorporates IslandPick , IslandPath-DIMOB and SIGI-HMM [70] . However , these results were discarded due to major incongruences between them . Finally , manual curation of all predicted MGEs was performed using Artemis v15 . 0 . 0 [71] ( S2 Table ) . In each of the six lineages , any recombined regions predicted by Gubbins that overlap with either repetitive regions or putative MGEs in the reference genome were discarded for the majority of the analysis in this study , leaving only putative homologous recombination regions . An in-house script was used to calculate the number of times each gene and each base had been involved in a homologous recombination event . Recombination “hotspots” were defined as genes with a recombination frequency above the 95th percentile observed in that particular ST and that have been involved in at least two events . Based on these criteria , the minimum number of recombination events that a gene must have been involved in to be considered within a hotspot region was four events in the ST1 lineage and two events in the remaining five STs . FastGEAR [34] was run on 54 individual gene alignments , comprising all 536 strains included in the study , which were extracted from the alignment of all genomes against the Paris reference . These genes span the prominent ST1 hotspot ( lpp1761-lpp1794 ) and also include 10 flanking loci on either side . The software infers the population structure of individual alignments , allowing detection of lineages in an alignment and “ancestral” and “recent” recombinations between them . The results were compared to those from Gubbins in terms of the number of recombination events predicted in each gene and the sharing of alleles among the different predicted lineages . Notably , if a recombination spans the entire length of an alignment , fastGEAR will detect this as another lineage in the alignment , rather than a recombination . Therefore , to make recombination counts between fastGEAR and Gubbins comparable , we used the estimated phylogeny and post-processed fastGEAR output by identifying branches in the tree where the population structure changed , and interpreted these as recombinations ( these can be seen as “blocks” with a colour different from yellow in S3B Fig and S4 Fig ) . The scripts used to make this calculation and to produce S3B Fig and S4 Fig can be found in https://users . ics . aalto . fi/~pemartti/fastGEAR/ . Homologous recombination regions were extracted from the ancestral sequences inferred from the nodes of the six phylogenetic trees , constructed prior to recombination removal , using PAML 4 [72] . Specifically , ancestral recombination sequences were extracted from the node downstream of the phylogenetic tree branch on which the recombination event was predicted to have occurred . A custom genome BLAST database ( BLAST v2 . 2 . 30+ ) [73] was constructed using de novo assemblies and/or complete genomes from all 536 L . pneumophila isolates used in this study . The reconstructed recombined regions were used as query sequences in BLAST searches against the custom genome database and the NCBI non-redundant nucleotide database . The resulting hits were filtered to remove those against isolates that are descended from the branch in which the recombination event was detected . Of the remaining hits , the BAPS cluster containing the isolate with the highest bit score was considered as the potential donor , provided that the hit covered at least 50% of the recombination fragment length and had a minimum of 99% nucleotide identity . Recombination fragments with no hits that met these thresholds were not assigned a donor cluster ( “No donor predicted” ) . Only recombined regions greater than 500bp were used in this analysis , firstly because they were deemed more likely to be a “true” event , and secondly because small regions would likely have high similarity to many genomes . | Legionella pneumophila is an environmental bacterium that causes Legionnaires’ disease , a serious and potentially fatal pneumonia . Previous studies have shown that members of this species undergo a process called recombination , whereby DNA is imported from another bacterial cell into the recipient genome . The imported DNA can either replace an equivalent segment of the genome ( homologous recombination ) or can comprise novel genes that are new to the recipient genome ( non-homologous recombination ) . Whilst recombination plays an undoubtedly important role in L . pneumophila evolution , accounting for more than 96% of the diversity observed within some lineages , little is known about its biological impact . In this study , we performed a detailed characterisation of the dynamics and effect of homologous recombination on L . pneumophila evolution in six clinically important lineages of L . pneumophila . We identified “hotspot” regions of the genome in which an excess of homologous recombination events was observed , which provided important clues to the selection pressures faced by L . pneumophila . By determining the donors of the recombined regions , we also revealed that recombination has occurred most frequently between isolates from the same clade , but also occurred between isolates from different major clades . This demonstrates the possibility of new adaptations arising in one lineage and being transferred to another distantly related lineage , which we predict has been an important factor in the emergence of several major disease-causing strains from diverse genomic backgrounds . | [
"Abstract",
"Introduction",
"Results",
"&",
"discussion",
"Materials",
"&",
"methods"
] | [
"taxonomy",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"microbiology",
"membrane",
"proteins",
"genomic",
"databases",
"phylogenetics",
"data",
"management",
"outer",
"membrane",
"proteins",
"phylogenetic",
"analysis",
"genome",
"analysis",
"dna",
"cellular",
"structures",
"and",
"organelles",
"bacteria",
"bacterial",
"pathogens",
"homologous",
"recombination",
"research",
"and",
"analysis",
"methods",
"genomic",
"libraries",
"legionella",
"pneumophila",
"computer",
"and",
"information",
"sciences",
"genomics",
"legionella",
"medical",
"microbiology",
"proteins",
"microbial",
"pathogens",
"biological",
"databases",
"recombinant",
"proteins",
"evolutionary",
"systematics",
"cell",
"membranes",
"biochemistry",
"cell",
"biology",
"nucleic",
"acids",
"database",
"and",
"informatics",
"methods",
"genetics",
"biology",
"and",
"life",
"sciences",
"dna",
"recombination",
"evolutionary",
"biology",
"computational",
"biology",
"organisms"
] | 2017 | Dynamics and impact of homologous recombination on the evolution of Legionella pneumophila |
Epstein-Barr virus ( EBV ) , an oncogenic human herpesvirus , induces cell proliferation after infection of resting B lymphocytes , its reservoir in vivo . The viral latent proteins are necessary for permanent B cell growth , but it is unknown whether they are sufficient . EBV was recently found to encode microRNAs ( miRNAs ) that are expressed in infected B cells and in some EBV-associated lymphomas . EBV miRNAs are grouped into two clusters located either adjacent to the BHRF1 gene or in introns contained within the viral BART transcripts . To understand the role of the BHRF1 miRNA cluster , we have constructed a virus mutant that lacks all its three members ( Δ123 ) and a revertant virus . Here we show that the B cell transforming capacity of the Δ123 EBV mutant is reduced by more than 20-fold , relative to wild type or revertant viruses . B cells exposed to the knock-out virus displayed slower growth , and exhibited a two-fold reduction in the percentage of cells entering the cell cycle S phase . Furthermore , they displayed higher latent gene expression levels and latent protein production than their wild type counterparts . Therefore , the BHRF1 miRNAs accelerate B cell expansion at lower latent gene expression levels . Thus , this miRNA cluster simultaneously enhances expansion of the virus reservoir and reduces the viral antigenic load , two features that have the potential to facilitate persistence of the virus in the infected host . Thus , the EBV BHRF1 miRNAs may represent new therapeutic targets for the treatment of some EBV-associated lymphomas .
Epstein-Barr virus ( EBV ) establishes a clinically silent chronic infection in the large majority of the world population [1] . In the course of primary infection , B cells targeted by the virus initiate transient cell growth until EBV-specific CD8+ T cells mount an efficient antiviral response [2] . However , some EBV-infected B cells will persist in the infected individual and form a long-lasting virus reservoir . In transplant recipients undergoing immunosuppressive therapy , EBV-induced B cell growth can lead to the development of a post-transplant lymphoproliferative disorder ( PTLD ) [1] , and analogous conditions have been observed in other immunodeficient patients . This process can be reproduced in vitro; exposure of resting B cells to the virus leads to the generation of continuously growing lymphoblastoid cell lines ( LCL ) . Both typical PTLD cells and LCLs produce all six Epstein-Barr virus nuclear antigens ( EBNA ) and three latent membrane proteins ( LMP ) , collectively designated latent proteins [1] . Viral mutants that lack EBNA1 , EBNA2 , EBNA-LP or LMP1 display a massive reduction in their ability to transform B cells [3]–[7] , thereby demonstrating that latent proteins are necessary for this process . However , it remains unclear whether they are sufficient for Bcell transformation . The EBV genome encodes a large number of non-coding RNAs that include the Epstein-Barr encoded RNAs ( EBERs ) , as well as 25 miRNAs and one small nucleolar RNA ( snoRNA ) [8]–[12] . MiRNAs bind to mRNAs that contain fully or partially complementary sequences and as a consequence usually impair their translation and reduce their stability [13] . The EBV miRNAs are distributed in two clusters , located in the BART region or within the BHRF1 gene locus [9] , [10] . The latter cluster comprises three members , two of which are located in the BHRF1 3′ untranslated region and the third immediately 5′ to the BHRF1 lytic mRNA transcription start site [9] , [10] ( Fig . 1A ) . They are processed from introns located within ∼100-kb long RNA transcripts that initiate at the Cp or Wp promoters used by all EBNA genes [14] . Expression of the BHRF1 miRNAs is characteristic of EBV-transformed B cells that express all latent genes ( latency III ) [1] , [10] . In contrast , Bcell tumours such as Burkitt's lymphomas or epithelial cell tumours that express a restricted number of latent proteins ( latency I or II ) do not express the BHRF1 miRNAs [10] . However , induction of virus replication in some Burkitt's lymphoma cells leads to re-expression of the miRNA BHRF1 cluster , suggesting that these miRNAs may serve functions not only during latency III but also upon induction of virus replication [15] . In an attempt to understand the function of the BHRF1 miRNA cluster in continuously growing lymphoblastoid B cell lines , we constructed viruses that lack these miRNAs and report here their phenotypic traits .
We deleted the BHRF1 miRNA cluster from the B95 . 8 genome cloned in E . coli in a sequential manner ( Fig . 1 ) . We first exchanged the miR-BHRF1-1 seed region for an unrelated sequence by chromosomal building using a shuttle plasmid that carries an ampicillin resistance cassette [16] . The modified miR-BHRF1-1 seed region introduces an AclI restriction site that allows unequivocal identification of the properly recombined mutant . In a second step , we replaced the DNA region that spans the miR-BHRF1-2 and miR-BHRF1-3 mature miRNAs by a kanamycin resistance cassette flanked by Flp recombinase target sites using recA-mediated homologous recombination . The last step consisted in excising the kanamycin resistance cassette by transient expression of the Flp recombinase . As a result , the miR-BHRF1-2 and miR-BHRF1-3 miRNAs were exchanged against one Flp recombinase target site ( Fig . 1 and Fig . S1 ) . The modified viral DNA , which carries a hygromycin resistance cassette , hereinafter referred to as Δ123 , was then transfected into 293 cells . Clones from these stably transfected 293 cells ( 293/Δ123 ) were obtained by hygromycin selection and the viral mutant genomes present in these producing cell lines were transferred back into E . coli and their global integrity was confirmed by restriction enzyme analysis ( Fig . 2A ) . Furthermore , sequencing the DNA fragments that were modified during virus construction confirmed the exactitude of the introduced alterations ( Fig . S2 ) and the complete identity of sequences outside the miRNAs with wild type genome . Next , the producer cell clones were tested for their ability to sustain viral lytic replication . The viral structural titers were detected by quantitative PCR and found to be similar to those observed with wild type producer cell lines . The mean values ranged between 2 . 2×107 and 2 . 9×107 genome equivalents per ml of supernatant for Δ123 and wt , respectively , showing that the BHRF1 miRNAs are not required for virus production ( Fig . 2B ) . We then incubated Raji B cells with these supernatants at various dilutions . Three days later the number of gfp-positive Raji cells was determined to assess functional infectious titers ( Fig . 2B ) . The ratio between structural titers ( geq/ml ) and functional titers ( gru/ml ) was found to be 7 . 8 and 10 . 3 for wt and Δ123 viruses , respectively . We therefore conclude , that the BHRF1 miRNAs are not essential for EBV infection but we cannot rule out a minor contribution to this process . The genome with the triple miRNA mutation formed the basis for construction of a revertant virus in which the modified sequences were re-exchanged with the original ones by chromosomal building to generate a Δ123 revertant ( Δ123 Rev ) virus genome [16] . This technique allows exact reconstruction of the original wild type sequence . The reverted genome was introduced in turn into 293 cells from which hygromycin-resistant clones were selected . Restriction analysis and sequencing confirmed that the revertant virus shared perfect homology with the wild type EBV genome ( Fig . 2A and Fig . S2 ) . Producer clones carrying the revertant genome delivered structural and functional titers akin to those observed with wt viruses ( Fig . 2B ) . To assess the contribution of the BHRF1 miRNA cluster to EBV's transforming properties , we exposed resting primary B cells to wild type , Δ123 , and Δ123 Rev viruses . Infections were carried out at an MOI of 1 infectious particle per B cell ( i . e . one gru/B cell ) , and cell outgrowth was monitored . Infected B cells were either seeded at low concentration , i . e . 2×103 per ml in a 96-well plate containing feeder cells or at high concentration i . e . 2×106 cells per ml . EBV-infected cells grow much more easily when infected at high concentration . Therefore , the first culture conditions are very stringent and allow detection of differences in terms of transformation efficiency but they do not allow monitoring of the infected B cells at the early stages of transformation . The percentage of wells containing outgrowing cell clones was assessed 8 weeks after infection . The results of three independent infection experiments is presented in Figure 3A . On average , wild type and revertant viruses respectively induced 82 and 75% cell outgrowth at an MOI of 1 . In contrast , only 3% of the wells containing B cells infected with Δ123 virus showed outgrowth ( Fig . 3A ) . Note that the standard variation between the different transformation assays was high . This reflects the fact that B cells from different individuals differ in their ability to form continuously growing cell lines . We conclude from these data that the BHRF1 miRNA cluster markedly increases the efficiency of transformation at low B cell concentration . The results of the bulk infection revealed similar though less pronounced effects . Monitoring of cell growth in B cells exposed to EBV-wt , Δ123 , and Δ123 Rev virus evidenced slower growth in samples infected with Δ123 . After 29 days in culture , B cells exposed to wild type or Δ123 Rev viruses expanded from 2×106 to 3 . 6×108 and 3 . 3×108 cells , respectively . This compares with 1 . 3×108 for B cells transformed by Δ123 ( Fig . 3B ) . This prompted us to study the cell cycle parameters of B cell populations , by performing a BrdU incorporation assay ( Fig . 3C ) . Whilst between 22 and 23% of B cells infected with both wild type controls entered S phase within 30 minutes , this percentage fell to 9 . 4% in B cells infected with Δ123 virus . We also noted that the ratio between cells still present in G2/M phase at the time of analysis and those that had entered S phase was ∼0 . 3 for the controls and ∼0 . 9 for B cells infected Δ123 virus . From this set of data , we conclude that the absence of BHRF1 miRNAs does not suppress EBV's transforming properties , but instead markedly slows down the growth rate of infected target cells . The permanently growing cell lines obtained after Δ123 infection provided material to quantify BHRF1 miRNA expression by RT-qPCR using the RNU48 snoRNA as an internal reference . The results of this analysis are given in Fig . 3D and show expression of the three BHRF1 miRNAs in all control lines tested , but not in those generated with Δ123 , as expected . Of note , the expression level of the BHRF1 miRNAs varies within the control LCLs established from three different B cell infections ( 1 to 5 range for miR-BHRF1-1 , 1 to 4 range for miR-BHRF1-2 , 1 to 2 range for miR-BHRF1-3 ) . This variation most likely reflects different activity levels of the Cp/Wp promoter among different LCLs [15] . The latent viral proteins have been recognized as the principal mediators of EBV's transforming properties . It was therefore important to assess latent gene expression in B cells infected with the EBV Δ123 mutant . To this aim , we carried out RT-qPCR analysis of the viral latent transcripts produced from the Wp and Cp promoters in transformed Bcell lines at 5 , 11 , 25 , 36 , and 73 days after infection ( Fig . 4A ) . At day 5 , the level of transcription from the Cp and Wp promoters was very similar in cells infected with Δ123 and in wild type controls . However , from day 11 on , the activity of the Wp promoter , and of the Cp promoter gradually increased in B cells transformed with the triple mutant relative to their control counterparts ( Fig . 4A ) . Rather than an absolute increase of Wp transcriptional activity in Δ123-positive B cell lines , the observed differences could be ascribed to a stronger down-regulation of Wp in the controls ( Fig . 4B ) . These results suggested that the BHRF1 miRNA cluster , directly or indirectly , negatively regulates the expression of Wp/Cp-driven transcripts . Alternatively , these results could be accounted for by selection of a minor population of B cells that happens to express very high levels of EBV latent genes , provided that such an overexpression confers a growth advantage in LCLs infected by EBV Δ123 mutant . To test this hypothesis , we performed clonality studies of the LCLs at the different time points used in transcriptional studies reported above ( Fig . 4C ) . To this aim , immunoglobulin heavy chain ( IgH ) genes were amplified by PCR using degenerated primers that bind to a large number of IgH variable gene families [17] . Whilst an IgH PCR performed on monoclonal B cell populations will reveal a unique band , it will yield a smear that ranges between approximately 330 and 350 bp if applied to a polyclonal B cell population as the length of different IgH family members slightly differs [17] . IgH PCR performed on LCLs generated with Δ123 or with wild type virus controls displayed a polyclonal pattern at day 0 ( before infection ) , day 5 , day 11 , and day 25 post-infection . From day 36 on , discrete bands become visible against a polyclonal background , and at day 73 all three cell lines contained multiple dominant discrete clones , ie they became oligoclonal ( Fig . 4C ) . From these results , we infer that early selection of dominant clones does not take place in the LCL infected by the EBV Δ123 mutant . We then used RT-qPCR and western blots to gauge EBV latent gene expression . The results of one out of four experiments are shown in Figure 5 . This set of experiments demonstrated that B cells generated with the Δ123 virus express all tested latent genes with some expressed at higher levels than in the controls at day 11 post-infection ( Fig . 5A ) . Similar assays conducted at day 36 post-infection confirmed this trend; all latent genes were expressed at higher levels in Δ123-positive LCLs both at the mRNA and the protein level ( Fig . 5B ) . This difference was particularly dramatic in the case of the EBNA-LP protein , which was five times more abundant in LCLs generated with Δ123 than in controls , but was also visible for EBNA1 , EBNA2 , EBNA3A , EBNA3B , and EBNA3C . LMP1 and LMP2-specific transcripts were also more abundant in LCLs infected by the triple mutant . Western blot analysis detected a marginal increase in LMP1 protein production in the same cells . Recent recognition that the BHRF1 protein is important for B cell transformation prompted us to assess expression of this gene [18] . Lymphoblastoid cell lines infected with wt , Δ123 , or Δ123 Rev viruses evinced similar levels of BHRF1 transcripts ( Fig . 5C ) .
Since their initial discovery in the EBV genome , miRNAs have been found in all investigated herpesviruses [19] , [20] . Although some of these miRNAs moderate the expression of virus replication proteins , e . g . miR BART2 and the EBV viral polymerase BART5 , miR-H2 , and the ICP0 transactivator , many have been found to regulate virus-cell interactions [21] , [22] . Maintenance of latency appears to be a recurrent theme across herpes viruses [20] , [23] . Examples are provided by HSV1 miR-M2-3p that inhibits translation of ICP0 , a key transactivator that initiates virus lytic replication [22] . More recently , a cluster of 12 pre-miRNAs from the KHSV genome has been implicated in the repression of lytic replication through its ability to down-regulate IκBα [24] . Facilitation of immune evasion appears to be another crucial function served by several virus miRNAs [20] . Indeed , miR-UL112-1 from HCMV curbs synthesis of the NK cellular cell surface receptor MICB and EBV miR-BHRF1-3 prevents production of the putative T cell chemoattractant CXCL11 [25] , [26] . In the present paper , we expand the spectrum of functions served by viral miRNAs during virus latency . We provide clear evidence that the BHRF1 miRNA cluster substantially enhances EBV's transforming potential . In its absence , B cells infected with the same functional MOI grew more slowly and could not efficiently form colonies at low concentration . Although we usually see EBV-mediated transformation as the molecular mechanism that underlies EBV's oncogenic properties , it is also activated during primary infection , presumably to expand the size of the Bcell reservoir . Our observations suggest that in the absence of the BHRF1 miRNA cluster , expansion of this compartment would be substantially reduced . Importantly , the Δ123 defective phenotype could not be ascribed to a down-regulation of latent protein expression . This family of proteins has been found to activate several signal transduction pathways that include the Notch , NFkB , Jun , AP1 , and JAK-Stat pathways and is therefore thought to be the principal effector of B cell transformation [14] . Although the level of latent transcription was nearly identical in B cells infected with Δ123 or in controls at day 5 post-infection , a divergence between both virus types became discernible after 11 days and became obvious after 3 weeks in culture . While B cells transformed with wild type viruses progressively but efficiently down-regulated Wp transcription starting at five days after infection , this process was much less efficient in cells transformed with Δ123 virus . In this respect , It is interesting to note that the kinetics of expression of the BHRF1 miRNA cluster and of Wp transcripts are exactly opposite; while miR-BHRF1-3 expression starts at day one and linearly increases until day 8 , Wp transcripts are maximal at day 1 and sharply decrease until day 8 after which they more slowly reach a minimum plateau at day 20 [27] , [28] . However , B cells infected by the Δ123 virus mutant also exhibited an increase in the abundance of Cp-initiated transcripts as soon as this promoter became dominant , three weeks after infection . Therefore , the deletion of the BHRF1 miRNA cluster prevents repression of the Wp transcripts and enhances expression of the Cp promoter . This could suggest that the latent transcripts , from which the latent genes originate , are direct targets of the miRNA BHRF1 cluster . However , the high amplitude of the effects caused by the absence of the BHRF1 miRNAs is not consonant with those usually observed with viral or cellular miRNAs , even if we hypothesize a synergistic effect of the three miRNAs . Rather , they suggest that the BHRF1 cluster may target one or more cellular transactivators , many of which ( BSAP/pax5 , RFX1 , YY1 , MIBP1 , CREB , ATF1 ) have been shown to regulate expression of the Wp/Cp promoters [14] . Using the photoactivatable ribonucleoside-enhanced crosslinking and immunoprecipitation ( PAR-CLIP ) technique recently described by the Tuschl laboratory [29] , we have comprehensively identified mRNAs bound by all of the viral and cellular miRNAs expressed in LCLs ( R . Skalsky and B . Cullen , manuscript in preparation ) . Using this technology , we have identified ∼300 mRNA targets for the EBV BHRF1 miRNAs . Although strong binding sites for cellular miRNAs on the viral LMP1 and BHRF1 mRNAs were readily detected , none of the latent EBV mRNAs were bound by any of the BHRF1 miRNAs . We therefore conclude that the observed effect of these viral miRNAs on EBV latent gene expression is indirect . The enhanced latent transcription observed in B cells infected with Δ123 virus correlated with an increased production of latent proteins relative to wild type controls . This effect was most visible for EBNA-LP , but EBNA1 , EBNA2 , EBNA3A , EBNA3B , EBNA3C , and LMP2A were all more strongly expressed in B cells infected with the triple miRNA mutant . Therefore , the total amount of viral antigens present in B cells transformed by Δ123 is substantially higher than seen in wild type controls . Latent proteins , and in particular EBNA1 , EBNA3A , and EBNA3C , have been shown to elicit a strong CD8+ T-cell response directed against EBV-infected cells . We infer from these data that the BHRF1 cluster reduces the antigenic load present in B cells and could therefore facilitate viral immune evasion [2] and thus persistence , a common trait of herpes viruses [20] . Is there a causal relationship between the relative excess of latent proteins observed in B cells transformed by Δ123 and the reduced growth rate ? An excess of LMP1 has been reported to exert toxic effects on infected B cells [3] , but the expression level obtained from the expression plasmid used in this work appears much higher than observed in Δ123-transformed LCLs . We therefore favor the view that the excess of latent genes cannot fully explain the Δ123 phenotype and that the BHRF1 cluster also acts independently of the viral latent genes , probably by targeting cellular genes involved in cell growth control . Our data clearly document that the BHRF1 miRNA cluster serves a central function in EBV biology as it enhances the virus' ability to dysregulate B cell growth . Thus , the miRNA cluster expands the size of EBV's reservoir and enhances its oncogenic property . Therefore , it represents a new potential therapeutic target for the treatment of EBV-associated diseases .
HEK293 cells are neuro-endocrine cells obtained by transformation of embryonic epithelial kidney cells with adenovirus [30] , [31] . Raji and Akata are EBV-positive Burkitt's lymphoma cell lines , BJAB is an EBV-negative Burkitt's lymphoma cell line [32]–[34] . WI38 are primary human lung embryonic fibroblasts [35] . All cell lines were grown in RPMI 1640 medium supplemented with 10% fetal calf serum ( FCS; Biochrom ) . We introduced mutations in miR-BHRF1-1 ( B95 . 8 co-ordinates 53762-53782; accession number V01555 . 2 ) by overlap PCR using mutated primers ( TAACCTGATCAGCCCC changed into TAACGTTGCAAGCCCC ) . This gave rise to a 1 . 8 kb fragment ( B95 . 8 coordinates 52807-54651 ) that encompasses a BHRF1-1 miRNA with a mutated seed region . Furthermore , the introduced mutation creates an AclI restriction site that allows distinction between the wild type and the mutated sequences ( see also Fig . 1 ) . The 1 . 8 kb PCR fragment was cloned into a plasmid ( B269 ) that consists of an arabinose-inducible temperature sensitive bacterial origin of replication , the ampicillin ( amp ) resistance gene , RecA and the LacZ operon to generate a shuttle vector ( B396 ) for chromosomal building . A 2 . 9 kb BspT1/SalI fragment ( B95 . 8 coordinates 53222-56081 ) was cloned into the vector B269 to generate the shuttle vector for construction of the Δ 23 revertant ( B 430 ) . The wild-type EBV B95 . 8 strain cloned onto a prokaryotic F-plasmid , which carries the chloramphenicol ( cam ) resistance gene , the gene for the green fluorescent protein ( GFP ) , and the hygromycin ( hyg ) resistance gene ( p2089 ) [36] , was used to generate the BHRF1 miRNA mutant . Deletion of miR-BHRF1-1 was performed by chromosomal building as described [16] using B396 as targeting vector . This involved initial building of a cointegrate between the shuttle vector that contains the mutated version of the mature miR-BHRF1-1 and the wild type EBV genome . This co-integrate that as a result of recombination carries both mutated and wild type miR-BHRF1-1 was then resolved to eliminate the wild type copy of the miRNA . Subsequently , a triple mutant lacking all three BHRF1 miRNAs was obtained by exchanging the viral DNA fragment that contains both mature miR-BHRF1-2 and miR-BHRF1-3 and the sequence between them ( B95 . 8 co-ordinates 55176-55279 ) against the kanamycin resistance gene flanked by flp-recombination sites as described [16] . The kanamycin resistance cassette from pCP15 was amplified using primers 682 CTTTTAAATTCTGTTGCAGCAGATAGCTGATACCCAATGTAACAGCTATGACCATGATTACGCC and 683 ATTTTAACGAAGAGCGTGAAGCACCGCTTGCAAATTACGTCCAGTCACGACGTTGTAAAACGAC . The kanamycin cassette was excised from recombined clones by transient transformation of the Flp recombinase cloned onto a temperature-sensitive plasmid ( pCP20 ) . A revertant clone for Δ123 was constructed by chromosomal building using the shuttle vector described above . Recombinant EBV plasmids were transfected into HEK293 cells by lipofection ( Metafectene , Biontex ) and selected for hygromycin resistance ( 100 µg/ml ) . Recombinant EBV genomes were purified from GFP-positive hygromycin-resistant cell clones as described [37] and electroporated into E . coli DH10B cells ( 1200 V , 25 mF , 200 Ω ) . The genetic integrity of the mutant or revertant EBV genomes stably introduced in HEK293 cell clones was assessed by restriction enzyme analysis and DNA sequencing of the BHRF1 gene region . The mutant producer cell clones were designated as 293/Δ123 and the revertant producer clones thereof as 293/Δ123 Rev . All described producer cell lines clones were lytically induced by transfection of a BZLF1 expression plasmid together with a BALF4 expression plasmid ( 3 µg each/6-well plate ) . Medium was changed at day 1 post-infection ( p . i . ) , virus supernatant harvested at day 4 p . i . , filtered through a 0 . 45 µm filter , and stored in aliquots at −80°C . Viral genome equivalents ( geq ) were determined by quantitative real-time PCR using EBV BALF5-specific primers and probe as described [38] . Infectious titers were determined by infecting 104 Raji B cells with increasing 5-fold dilutions of supernatants . Three days after infection , gfp-positive cells were counted using a fluorescent microscope . Primary B cells were freshly isolated from adult human blood buffy coats by density gradient centrifugation followed by positive selection using M-450 CD19 Dynabeads ( Dynal ) . To quantify EBV-mediated B cell transformation rates , freshly isolated primary B cells were exposed to infectious supernatants at a MOI of 1 infectious particle ( gru ) per B cell overnight at 37°C and then seeded at a concentration of 100 cells per well in 96-U-well plates coated with lethally irradiated WI38 feeder cells . The number of wells with proliferating cells was determined 8 weeks post-infection . Infection experiments were also carried out at high B cell concentration; 2x10e6 B cells were exposed to infectious supernatant at an MOI of 1 infectious particle per cell and were harvested for RNA and protein extraction at different time points post-infection . BHRF1 miRNAs extracted from lymphoblastoid cell clones were reverse transcribed using specific stem-loop primers and TaqMan miRNA reverse Transcription Kit ( Applied Biosystems ) as described elsewhere [28] , [39] . In brief , 110 ng of total RNA isolated with a miRNA extraction kit ( Qiagen ) was reverse transcribed as recommended by the manufacturer . The sequences of stem-loop primer , primer and probes and the qPCR conditions were identical to those described elsewhere [28] . Reverse transcription and amplification of cellular snoRNA RNU48 was performed in parallel to normalize for cDNA recovery ( Assay ID 001006; Applied Biosystems ) as recommended by the manufacturer . Real-time PCR was performed on an ABI 7300 Sequence Detection System ( Applied Biosystems ) . All reactions were run in duplicates . An aliquot of the PCR reaction was taken after 30 cycles , products were loaded onto a 8% polyacrylamide gel and visualized by ethidium bromide staining . Cells were resuspended in PBS and lysed by sonication . 30 µg of proteins were denatured in Laemmli buffer for 5 minutes at 95°C , separated on a 10% or 7 . 5% SDS-polyacrylamide gel and electroblotted onto a Hybond ECL membrane ( Amersham ) . After preincubation for 30 min in 5% milk powder in PBS , blots were incubated with primary antibodies EBNA1 ( IH4 ) , EBNA2 ( PE2 ) , LMP1 ( S12 ) , EBNA-LP ( JF186 ) , EBNA3A ( ExAlpha ) , EBNA3B ( ExAlpha ) , EBNA3C ( A10 ) , or actin ( clone ACTN05; Dianova ) for 1 h at room temperature . After several washings in 0 . 1% Tween in PBS , blots were incubated for 1 h with secondary antibodies coupled with horseradish peroxidase . Antibody binding was revealed using an ECL detection reagent ( Perkin Elmer ) . Densitometric analyses of immunoblots were performed using ImageJ software . 400 ng aliquots of RNA isolated from infected B cells ( miRNeasy kit , Qiagen ) was reverse transcribed with AMV-RT ( Roche ) using a mix of primers specific for various EBV transcripts and specific for GAPDH as described [18] , [40] . Quantitative PCR using primers specific for Cp , Wp , LMP1 , LMP2A , EBNA2 , and YUK-spliced EBNA1 transcripts were performed as described [40] . PCR and data analysis was carried out using the universal thermal cycling protocol on an Applied Biosystem 7300 real-time PCR System . All samples were run in duplicates , together with primers for the amplification of the human GAPDH gene in combination with a VIC-labeled probe ( Applied Biosystems ) to normalize for variations in cDNA recovery . Cell cycle parameters of exponentially growing lymphoblastoid cell lines phase were monitored by exposing 1×106cells for 30 minutes to BrdU ( 10 µM final concentration ) ( BrdU flow kit , BD Biosciences Pharmingen , San Diego , USA ) . The degree of incorporation was determined by immunostaining with an APC-coupled antibody directed against BrdU . DNA content was assessed by staining cells with the DNA intercalating dye 7-AAD . Analysis of fluorescence with a two laser FACScalibur flow cytometer allows identification of cells that have entered the S phase ( BrdU positive ) , the G2/M phase ( BrdU negative , double DNA content ) or the G0/G1 phase ( BrdU negative , double DNA content ) . DNA was extracted from 1 to 3×106 cells using a DNeasy Tissue kit ( Qiagen ) . IgH variable ( IgVH ) sequences spanning FR1 , CDR1 , FR2 , CDR2 , FR3 , and CDR3 were PCR-amplified using a single consensus forward primer FR1c:5′AGGTGCAGCTGSWGSAGTCDGG 3′ and a mixture of JH family-specific reverse primers JH1/2/4/5 5′-ACCTGAGGAGACGGTGACCAGGGT-3′ , JH3 5′- TACCTGAAGAGACGGTGACCATTGT-3′ and JH6 5′- ACCTGAGGAGACGGTGACCGTGGT-3′ described previously [17] . PCR amplification was performed in a reaction containing 1 µg heat denatured ( 10 min at 94°C ) genomic DNA , 0 . 8 µM FR1c primer , 0 . 26 µM of each JH primer , 200 µM of each dNTP , 1 . 0 mM MgCl2 , and 2 . 5 U Expand High Fidelity thermostable DNA polymerase ( Roche ) . The first PCR cycle consisted of a denaturing step at 94°C for 2 min , followed by 30 cycles at 94°C for 30 s , at 61°C for 60 s , and at 72°C for 60 s ( 10 min in the last cycle ) . PCR products were then separated by electrophoresis on a 6% polyacrylamide gel , and bands corresponding to the 330–350 bp IgVH products visualized by ethidiumbromide staining . | To persist in their hosts , herpes viruses must avoid recognition by the host's immune system . Down-regulation of viral antigen production through expression of viral miRNAs is a particularly elegant mechanism as these genetic elements do not encode proteins and remain therefore invisible to the immune system . Upon primary infection , Epstein-Barr virus ( EBV ) colonizes B cells and , through expression of its latent proteins , induces their continuous proliferation . The resulting expansion of infected B cells elicits a T cell response directed against the latent proteins that results in their elimination . Therefore , rapid proliferation of infected B cells , combined with reduced latent protein production , would facilitate establishment of EBV's viral reservoir before mounting of the immune response . Here , we find that a cluster of three microRNAs encoded near the EBV BHRF1 gene is crucial for efficient B cell transformation . In the absence of these genetic elements , infected B cells grow markedly more slowly . Furthermore , B cells exposed to an EBV mutant that lacks the BHRF1 microRNA cluster produced more latent proteins . Thus , the BHRF1 microRNA cluster possesses properties that potentiate EBV's oncogenic properties and therefore facilitate expansion of the EBV B cell reservoir . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"virology/persistence",
"and",
"latency",
"virology/viruses",
"and",
"cancer",
"virology"
] | 2011 | A Viral microRNA Cluster Strongly Potentiates the Transforming Properties of a Human Herpesvirus |
Tiller angle is one of the most important components of the ideal plant architecture that can greatly enhance rice grain yield . Understanding the genetic basis of tiller angle and mining favorable alleles will be helpful for breeding new plant-type varieties . Here , we performed genome-wide association studies ( GWAS ) to identify genes controlling tiller angle using 529 diverse accessions of Oryza sativa including 295 indica and 156 japonica accessions in two environments . We identified 7 common quantitative trait loci ( QTLs ) , including the previously reported major gene Tiller Angle Control 1 ( TAC1 ) , in the two environments , 10 and 13 unique QTLs in Hainan and Wuhan , respectively . More QTLs were identified in indica than in japonica , and three major QTLs ( qTA3 , qTA1b/DWARF2 ( D2 ) and qTA9c/TAC1 ) were fixed in japonica but segregating in indica , which explained the wider variation observed in indica compared with that in japonica . No common QTLs were identified between the indica and japonica subpopulations . Mutant analysis for the candidate gene of qTA3 on chromosome 3 indicated a novel gene , Tiller Angle Control 3 ( TAC3 ) , encoding a conserved hypothetical protein controlling tiller angle . TAC3 is preferentially expressed in the tiller base . The ebisu dwarf ( d2 ) mutant exhibited a decreased tiller angle , in addition to its previously described abnormal phenotype . A nucleotide diversity analysis revealed that TAC3 , D2 and TAC1 have been subjected to selection during japonica domestication . A haplotype analysis identified favorable alleles of TAC3 , D2 and TAC1 , which may be used for breeding plants with an ideal architecture . In conclusion , there is a diverse genetic basis for tiller angle between the two subpopulations , and it is the novel gene TAC3 together with TAC1 , D2 , and other newly identified genes in this study that controls tiller angle in rice cultivars .
As a key factor in rice plant architecture , tiller angle determines the planting density per unit area and contributes greatly to grain yield . The tiller angle is defined as the angle between the main culm and the side tillers [1] . A favorable rice tiller angle is an important component of the ideal plant architecture that has been selected by humans in the long history of domestication and genetic improvement [2] . Oryza rufipogon , the progenitor of cultivated rice ( Oryza sativa ) , shows a spread-out growth pattern that allows it to escape some diseases induced by high humidity , but this pattern occupies too much space and increases shading and lodging , thus decreasing the photosynthetic efficiency and grain yield per unit area accordingly . By contrast , cultivated rice usually exhibits a better plant architecture , with a smaller tiller angle that leads to high potential yields . Tiller angle changes dynamically throughout the life cycle of rice to achieve efficient resource use . At tillering stage , tiller angle increases , which enables young plant to occupy a large empty space for subsequent tiller development and simultaneously inhibits weed growth . At later stages , tiller angle decreases , especially at heading stage , when it reaches its minimum , allowing the mature plant to reduce leaf shade and increase its photosynthetic efficiency . Heading stage is a key developmental period in rice , and the planting density per unit area is mainly determined by the tiller angle at this stage . Due to its importance in rice production , increasing attention has been paid to dissecting the genetic basis of tiller angle over recent decades . Dozens of tiller angle-related quantitative trait loci ( QTLs ) have been explored in rice via classical bi-parental cross mapping [3–8] . However , most of these QTLs are seldom used for marker-aided selection , due mainly to their minor effects . Recently , a few genes controlling tiller angle have been cloned . PROSTRATE GROWTH 1 ( PROG1 ) has been accepted as a domestication-related gene that controls tiller angle and tiller number during both tillering and heading stages in wild rice . PROG1 encodes a zinc finger nuclear transcription factor and is located on chromosome 7 . Amino acid changes in the PROG1 protein and regulatory changes during domestication led to loss of the function of this gene , which promoted the transition from the prostrate growth tillers of wild rice to erect growth tiller of domesticated rice [9 , 10] . Tiller Angle Control 1 ( TAC1 ) is a major QTL located on chromosome 9 that controls tiller angle during heading stage in cultivated rice and encodes an expressed protein without homologous genes in rice . A mutation in the 3’-splicing site of the 1 . 5-kb intron ‘GGGA’ , which exists in 88 compact japonica rice accessions , decreases the level of tac1 and leads to a smaller tiller angle; ‘AGGA’ is present in 21 wild rice and 43 indica rice accessions with spread-out tillers [11] . Rice shoot gravitropism is suggested to be a key factor affecting plant architecture . The rice la mutant , which exhibits a wider tiller angle , has been intensively studied for decades [12–17] . However , the LAZY1 ( LA1 ) gene was not identified until 2007 . LA1 encodes a novel , grass-specific protein that controls shoot gravitropism by regulating polar auxin transport ( PAT ) [18 , 19] . Loose Plant Architecture 1 ( LPA1 ) is an INDETERMINATE DOMAIN protein involved in shoot gravitropism that regulates both tiller angle and leaf angle in rice during the vegetative and reproductive stages [20] . Suppression of OsPIN1 or over-expression of OsPIN2 ( two auxin efflux transporters ) alters PAT and increases tiller angle [21 , 22] . Recent research has demonstrated that strigalactones ( SLs ) , a group of newly identified plant hormones , inhibit auxin biosynthesis and attenuate rice shoot gravitropism primarily by decreasing local indoleacetic acid contents . Multiple SOLs ( suppressors of lazy1 ) involved in SL biosynthesis , such as dwarf 17 ( d17 ) , d10 and d27 , or in SL signaling pathways , such as d14 and d3 , can rescue the spreading phenotype of lazy1 [23] . Suppressing the expression of OsLIC1 ( Oryza sativa leaf and tiller angle increased controller ) , a novel CCCH-type zinc finger gene , increases leaf and tiller angles via the BR signaling pathway [24] . Although the identification of these tiller angle-related genes is helpful for understanding the mechanism of tiller angle formation , few genes that can be used for improving rice plant architecture have been isolated based on natural variation . Therefore , mapping additional genes that contribute to the natural variation of tiller angle is required for breeding varieties with an ideal plant architecture resulting in high grain yields . Genome-wide association studies ( GWAS ) offer a powerful approach to establishing the relationship between DNA markers and phenotypic traits in crops [25] . With the assessment of millions of single nucleotide polymorphisms ( SNPs ) , GWAS can take full advantage of ancient recombination events to identify the genetic loci underlying complex traits at a high resolution using a large number of crop varieties [26] . Many QTLs for agronomic traits have been identified in cultivated rice through GWAS [27–29] . Recently , several novel loci and candidate genes for tiller angle at tillering stage have been identified by GWAS and the elite alleles have been explored for plant architecture improvement [30] . Here , we investigated the tiller angle of 529 O . sativa accessions at heading stage and performed GWAS separately in the full population and the indica and japonica subpopulations . We isolated a novel gene , TAC3 , as well as several novel QTLs in this study . We also identified distinct genetic regulatory mechanisms for tiller angle between the two subpopulations , providing information on how to improve tiller angle in indica and japonica rice .
The worldwide rice collection exhibited a distinctive population structure and was classified into nine subpopulations: indI , indII , indica intermediate , Tej , Trj , japonica intermediate , Aus , VI , and intermediate [31] . Of these 529 accessions , 295 were classified into the indica subpopulation , including indI , indII and indica intermediate , and 156 were classified into the japonica subpopulation , including Tej , Trj and japonica intermediate . There were large variations in tiller angle throughout the population in both environments . The tiller angle ranged from 2 . 5° to 34 . 4° in Hainan and from 1 . 8° to 31 . 5° in Wuhan . The largest number of accessions fell into a small range of tiller angles , from 2° to 16° ( Fig 1a ) . The variations observed in the two environments showed a similar distribution , skewed towards smaller tiller angles . A significant correlation was observed between the two environments in the whole population ( r = 0 . 66 ) . However , there was a significant difference in tiller angle between the indica and japonica subpopulations ( Fig 1b and 1c ) . On average , indica rice exhibits a larger tiller angle ( 11 . 7 ± 5 . 8° in Hainan; 10 . 5 ± 5 . 6°in Wuhan ) than japonica rice ( 8 . 8 ± 3 . 6° in Hainan; 9 . 1 ± 3 . 6° in Wuhan ) , and the variation within indica is greater than in japonica . The correlation coefficients between Hainan and Wuhan were 0 . 71 and 0 . 49 within the indica and japonica subpopulations , respectively . Two-way analysis of variance ( ANOVA ) revealed that tiller angle was dominantly controlled by genetic factors but was also influenced by interactions between genotype and environment ( Table 1 ) . In the japonica subpopulation in particular , the interaction between genotype and environment accounted for 22 . 6% of the variation in tiller angle ( Table 1 ) . Tiller angle had a high heritability of 0 . 82 . We performed GWAS separately in the whole population and in the indica and japonica subpopulations for each year . Manhattan plots and quantile-quantile plots of the rice tiller angles among the three populations are illustrated with the results obtained from both the linear mixture method ( LMM ) ( S1 Fig ) and linear regression ( LR ) approaches ( S2 Fig ) . A total of 30 tiller angle-related QTLs were detected ( Table 2 and S1 Table and S2 Table ) . Of them , seven were commonly detected in Wuhan and Hainan . Three QTLs ( qTA1b , 3 and 7a ) were only identified through LR , while the remaining was identified through both LMM and LR ( Table 2 ) . Among these seven QTLs , 3 ( qTA1b , 7a and 8b ) were detected only in the whole population , 2 ( qTA3 and 7b ) were identified only in the indica subpopulation , and two QTLs ( qTA8a and 9c ) were commonly detected in the whole population and the indica subpopulation . No significant association signals were commonly detected in the japonica subpopulation in the two environments ( Table 2 ) . Two QTLs ( qTA1b and 3 ) on chromosomes 1 and 3 individually explained more than approximately 15% of the variation in the whole population and in the indica subpopulation , respectively . Two QTLs ( qTA8a and 8b ) on chromosome 8 presented different contributions to the tiller angle in the two environments . The QTL of qTA8a exhibited the major effect in the indica subpopulation . In addition to these 7 QTLs commonly detected in both environments , 10 and 13 QTLs were specifically detected in Hainan ( S1 Table ) and Wuhan ( S2 Table ) , respectively . In Hainan , 5 , 4 and 3 QTLs were detected in the whole population and in the indica and japonica subpopulations , respectively . Two QTLs ( qTA7d and qTA12a ) were commonly detected in the whole population and in the indica subpopulation . Three QTLs were located on chromosome 7 . In Wuhan , 4 , 3 and 6 QTLs were detected in the whole population and in the indica and japonica subpopulations , respectively . However , no QTLs were commonly detected in either population . There were 3 QTLs on each of chromosomes 1 and 7 . In the past decade , ~11 genes were reported to control tiller angle in rice [32] . However , in the present study , only TAC1 was detected in the local LD region via GWAS in both Hainan and Wuhan ( Table 2 ) . To evaluate the results of GWAS , we compared the localization of associated sites with those 11 tiller angle QTLs detected in cultivated rice from the gramene web site ( http://www . gramene . org ) and 14 significant tiller angle loci detected via GWAS in the previous study[30] . A total of 7 associated sites were found to co-localize with 6 previously reported QTLs ( S3 Table ) . Of them , four QTLs were commonly identified in both environments . qTA1b and qTA9c were found in the regions of QTa1 and qTA-9a , respectively; Both qTA8a and qTA8b were located in a large QTL interval covered a 4-Mb genome region [8 , 30] . qTA4 and qTA7h detected only in Wuhan and qTA9b detected in Hainan were identified in QTL regions . The co-localization of QTLs detected through linkage analysis and GWAS indicated their roles in controlling tiller angle . The lead SNP sf0329582676 of qTA3 on chromosome 3 , which was specifically identified in indica rice , is located in the second intron of Os03g51670 , 5 kb upstream of Os03g51660 ( Fig 2b ) . We quantified all 49 SNPs between Os03g51660 and Os03g51670 in 295 indica accessions , and a representation of the obtained pairwise r2 values showed that some SNPs in both genes were in high linkage disequilibrium ( LD ) with each other . SNPs in the 5’ end of Os03g51660 and Os03g51670 were grouped into one LD block , which made it difficult to determine the candidate gene of qTA3 ( Fig 2a ) . A mutant ( 05Z11AZ62 ) was subsequently obtained in which a T-DNA was inserted in the 3’UTR ( 441 bp from stop codon ) of Os03g51670 and the promoter ( 991 bp from start codon ) of Os03g51660 ( Fig 2b and 2c ) ; this mutant showed a larger tiller angle compared with the wild type at both tillering and heading stages in 2015 and 2016 of Wuhan ( Fig 2g–2i ) . A segregation analysis of a family of 96 plants ( chi square = 3 . 56 ) indicated that a single gene controlled the tiller angle . The average tiller angle of homozygous mutants was the largest , while that of heterozygous mutants was intermediate , and wild type plants possessed the smallest tiller angle at both stages in two years ( Fig 2g ) . An expression profiling analysis indicated that Os03g51660 was highly expressed in the tiller base , and the constitutive expression of Os03g51670 ( S3d and S3e Fig ) implied that Os03g51660 was the candidate gene controlling the tiller angle . We investigated the expression level of Os03g51670 and Os03g51660 between mutants and wild type . The expression level of Os03g51670 was significantly reduced only in the homozygous mutant and not in the heterozygote compared with wild type ( Fig 2f ) . However , the expression level of Os03g51660 was significantly increased in the homozygous mutants and the heterozygotes compared with wild type ( Fig 2e ) . That is , the Os03g51660 expression level co-segregated with the tiller angle . In addition , Os03g51660 was preferentially expressed in the tiller base , where the tiller bud is initiated and outgrown . Therefore , Os03g51660 is the gene underlying qTA3 . Hereafter , this gene is referred to as TAC3 , and this mutant is named as tac3-1D . To provide strong evidence for our results , another two mutants 1B-24636 ( tac3-2D ) with a T-DNA insertion in the 5’UTR ( 146 bp from start codon ) of Os03g51660 and 4A-02006 with a T-DNA insertion in the first intron of Os03g51670 ( Fig 2b ) were investigated in 2016 . The expression level of Os03g51660 in both homozygous and heterozygous tac3-2D mutants significantly increased compared with that in wild type Dongjing ( Fig 2j ) . The average tiller angle was no significant difference between heterozygous tac3-2D and wild type Dongjing at both stages , but the tiller angle of homozygous mutant ( 12 . 3°±2 . 9° tillering stage and 11 . 4°±3 . 8° heading stage ) was significantly larger up to 3° than DJ ( 9 . 0°±2 . 9° tillering stage and 8 . 6°±2 . 0° heading stage ) ( Fig 2l–2n ) . Whereas the expression of Os03g51670 was not changed in tac3-2D as compared to wild type ( Fig 2k ) . Although expression of Os03g51670 was significantly decreased in both homozygous and heterozygous 4A-02006 mutants ( S3a and S3b Fig ) , tiller angle of mutants was not significantly changed ( S3c Fig ) . Therefore Os03g51660 is the identity of TAC3 controlling tiller angle . The genomic DNA sequence of TAC3 is 1 , 717 bp in length , with five exons and four introns ( Fig 2b ) . Its coding sequence is 459 bp in length ( S3f Fig ) , and it encodes a conserved hypothetical protein of 152 amino acids ( http://rice . plantbiology . msu . edu/ ) . We detected 3 major haplotypes based on 10 SNPs ( Minor Allele Frequency ( MAF ) ≥ 0 . 05 ) in TAC3 among the 295 indica accessions ( Table 3 ) . The tiller angle of Hap1 was significantly wider than that of Hap2 and Hap3 in both environments ( p≤0 . 01 ) . However , only 1 G/T SNP ( sf0329577726 ) was found in the japonica subpopulation , and all but two accessions carried the ‘T’ allele , which indicated that a major allele dominated in japonica; by contrast , the ‘G’ allele was present in all but 10 indica accessions . Nucleotide diversity analysis showed that both TAC3 ( π = 1 . 9e-5 ) and its surrounding genomic region ( π = 1 . 1e-4 ) of 100 kb upstream and downstream presented significantly decreased values compared with the average nucleotide diversity across the whole japonica genome ( π = 1 . 45e-3 ) , indicating that TAC3 was selected during japonica domestication and genetic improvement ( Table 4 ) . qTA9c was closely linked to TAC1 , a previously reported rice tiller angle-related gene [11] , and its lead SNP was located 1 . 4 kb upstream to TAC1 . Assessment of the LD between the lead SNPs and all polymorphic sites in TAC1 showed that the lead SNP sf0920735688 ( PLMM = 4 . 2e-14 ) was in high linkage disequilibrium with most polymorphic sites in TAC1 containing GGGA/AGGA ( sf0920731363 ) sites , which are known functional nucleotide polymorphic sites ( FNPs ) of TAC1 ( Fig 3 ) . Therefore , TAC1 was assumed to be the candidate gene of qTA9c . We further investigated a total of 13 SNPs ( MAF≥0 . 05 ) throughout TAC1 among 529 O . sativa accessions , and 16 haplotypes were detected . The vast majority ( 285 indica and 156 japonica accessions ) of the accessions belonged to Hap1 , Hap2 and Hap3 ( Table 5 ) . All but one of the japonica accessions were classified into Hap3 with the ‘GGGA’ allele . Most of the indica accessions fell into Hap1 and Hap2 , which both contain the ‘AGGA’ allele , whereas 35 indica accessions carried Hap3 . There was a highly significant difference ( p<0 . 01 ) in the tiller angle between Hap1 and Hap3 in the indica subpopulation in both environments ( Table 5 ) . These results explained why TAC1 was detected in the whole population and the indica subpopulation , but not in the japonica subpopulation . The nucleotide diversity of TAC1 and its flanking region was subsequently analyzed . Accordingly , a significant reduction in nucleotide diversity was observed at TAC1 in the japonica subpopulation ( π = 9 . 4e-5 ) compared with both wild rice ( π = 1 . 45e-3 ) and the average diversity of the whole japonica genome ( π = 1 . 45e-3 ) ( Table 4 ) . We also observed a lower level of nucleotide diversity ( π = 1 . 7e-4 ) in the 100-kb region surrounding TAC1 in the japonica subpopulation compared with the whole japonica genome . However , no obvious changes were found in the indica subpopulation ( π = 3 . 0e-3 ) . Brassinosteroid , as one of the main plant hormones , is considered to be an important factor in plant development , including the tiller angle . The lead SNP sf0105241388 of qTA1b is located in the last intron of DWARF 2 ( D2 ) . A representation of the obtained pairwise r2 values showed that all SNPs in D2 were in one LD block ( Fig 4a ) . The d2-1 and d2-2 mutant shows a brassinosteroid-deficient phenotype including an erect leaf angle and a short plant height , but the tiller angle of this mutant was not mentioned [33] . Thus , the tiller angle of a mutant we named d2-3 ( 04Z11MY27 ) with a T-DNA insertion in the third intron of D2 was investigated ( Fig 4b ) . Genotyping ( Fig 4c ) and an expression analysis ( Fig 4d ) suggested that the d2-3 mutant exhibited abnormal growth , including a smaller tiller angle ( Fig 4e–4g ) . This result suggested D2 as the candidate gene of qTA1b . We further investigated the tiller angle of d2-1 and d2-2 to solid our study , and found the tiller angles of both mutants ( d2-1: 5 . 2°±1 . 3° tillering stage and 3 . 6°±1 . 2° heading stage; d2-2: 6 . 3°±1 . 5° tillering stage and 3 . 8°±1 . 2° heading stage ) were significantly decreased as compared with that in wild type T65 ( 11 . 3°±3 . 1° tillering stage and 7 . 6°±1 . 5° heading stage ) ( Fig 4h–4j ) . To better understand the natural variation of D2 , we further analyzed its haplotypes . We constructed the haplotypes of D2 within two subpopulations based on the SNPs with an MAF ≥ 0 . 05 . In the japonica subpopulation , we obtained four main haplotypes excluding the lead SNP sf0105241388 because all but one accession carried the same allele , ‘A’ , at this site . Most of the accessions ( 112 ) fell into one haplotype . Only 13 , 11 and 10 accessions fell into the other three haplotypes . The tiller angle showed no significant differences among these haplotypes ( Table 6 ) . In the indica subpopulation , we also obtained four main haplotypes that were different from those in the japonica subpopulation ( Table 7 ) . Hap1 and Hap4 exhibited significantly smaller tiller angles compared with Hap2 and Hap3 ( p≤0 . 01 ) . Hap1 and Hap4 carried the allele ‘A’ at the lead SNP site , whereas Hap2 and Hap3 carried the ‘G’ allele . Significantly decreased nucleotide diversity ( π = 1 . 6e-4 ) was observed only in the japonica subpopulation compared with the average nucleotide diversity of its whole genome ( Table 4 ) . According to the results described above , it is clear that TAC3 , D2 and TAC1 together , along with other genes , contribute to the natural variation in tiller angle in cultivated rice . Thus , the combinations of TAC3 , D2 and TAC1 would be expected to result in wide variation of tiller angles in cultivars . A total of 77 combinations of the genes TAC3 , D2 and TAC1 were observed in the 295 indica accessions , but only 9 combinations that were each found in more than 10 accessions were used for the comparative analysis ( Table 8 ) . In Wuhan and Hainan , combinations 1 and 2 both pyramided three haplotypes that increased the tiller angle , resulting in the largest tiller angle . Combinations 3–7 with decreased tiller angle haplotypes of D2 and TAC3 , independent of TAC1 showed similar compact plant architecture in both environments . Combination 8 pyramided increasing tiller angle haplotypes of D2 and TAC3 , but a decreased tiller angle haplotype of TAC1 resulted in a large tiller angle . Combination 9 carried increasing tiller angle haplotypes of TAC1 and TAC3 but a decreasing tiller angle haplotype of D2 , and accordingly exhibited a large tiller angle as well . These results indicated that D2 and TAC3 were the major QTLs in the indica cultivars . Among all 295 indica accessions , 5 carried two combinations with decreasing tiller angle haplotypes at three gene loci . Accessions W066 and W155 ( TAC3-Hap3/D2-Hap1/TAC1-Hap3 ) showed more compact tiller angles ( W066: 3 . 8° in Hainan and 3 . 5° in Wuhan; W155: 6 . 0° in Hainan and 5 . 5° in Wuhan ) . Similarly , accessions W161 , W243 and W244 ( TAC3-Hap2/D2-Hap1/TAC1-Hap3 ) also presented smaller tiller angles ( 5 . 2–6 . 7° in Hainan and 5 . 2–5 . 7° in Wuhan ) .
In general , indica rice exhibits a wide plant type , and japonica rice exhibits a compact plant type . In this study , the indica subpopulation presented wider variation in tiller angle compared with the japonica subpopulation ( Fig 1 ) . Although genetic effects explained more of the variation in tiller angle ( Table 1 ) , Genotype by environment interactions also significantly affected tiller angle . Moreover , genetic factors contributed more to the variation in tiller angle in indica than in japonica . These results indicated that there are probably different genetic bases underlying the tiller angle in the two subpopulations . Coincidently , GWAS identified more QTLs for the tiller angle in the indica subpopulation than in the japonica subpopulation in both environments . This result indicates that the greater number of QTLs controlling the tiller angle in indica rice would contribute to wider natural variation . Although a dozen QTLs were identified in each subpopulation , there were no common QTLs detected in both subpopulations in this study . Two QTLs , qTA8a and qTA9c , were found in both the whole population and the indica subpopulation , but not in the japonica subpopulation . It is likely that both QTLs were fixed for one major haplotype in the japonica accessions , resulting in failed detection . For example , the locus qTA9c/TAC1 was not identified in japonica because TAC1 is dominantly fixed with Hap3 in this subpopulation ( Table 5 ) . The tiller angle associated with Hap3 was significantly reduced compared with that of Hap1 , the major haplotype in indica , which explains the compact plant architecture of japonica rice . There was only one SNP observed in qTA3/TAC3 in japonica , and all but two accessions carried the dominant allele . Thus , the japonica accessions were almost fixed with one haplotype for TAC3 . Accordingly , the haplotype analysis of the full population suggested the existence of indica-japonica differentiation for D2: no significant differences in tiller angle were observed among the four main haplotypes within the japonica subpopulation ( Table 6 ) , while there was a significant difference in the indica subpopulation ( p≤0 . 01 ) ( Table 7 ) . This suggested that the major QTLs are fixed in japonica . Two QTLs , qTA7a and qTA8b , were only identified in the whole population and were not identified in either the indica or japonica subpopulation . A possible explanation for this finding is that the gene was fixed with different haplotypes in the two subpopulations . When GWAS was performed in each subpopulation , the QTLs were not detected because no polymorphism was present in sequence; however , when GWAS was conducted in the whole population , these QTLs were identified because variation occurred . Hence , it is recommended that GWAS should be performed separately in different populations to discover more QTLs for tiller angle or other traits . According to the above results , we propose that there is diverse genetic basis controlling the tiller angle between the two subpopulations . Tiller angle is a domestication-related trait in rice , and selective signatures of domestication include a reduction of nucleotide diversity and altered allele frequencies at domestication-related loci [34 , 35] . The functional SNPs of PROG1 , a gene controlling prostrate growth , only exist in wild rice [9 , 10] . In addition , PROG1 was detected via GWAS in 446 O . rufipogon accessions and was screened for strong selection signals [36] . In this study , PROG1 was not identified in the examined cultivars . LA1 is another major gene isolated from a mutant that controls the tiller angle , and loss of function of LA1 destroys rice shoot gravitropsim through altering the polar transport of auxin [19] . The lazy1 mutant shows a wider tiller angle of more than 60° , especially during the mature stage , associated with a prostrate growth phenotype [18] , which was never observed in our collection . Therefore , la1 and PROG1 do not contribute to the variation of tiller angle in our cultivars . In this study , GWAS and a mutant analysis demonstrated that TAC3 regulates tiller angle . The nucleotide diversity of TAC3 in japonica ( π = 1 . 9e-5 ) was approximately 55-fold lower than that of 111 randomly chosen gene fragments from japonica ( π = 1 . 1e-3 ) [37] , which is equivalent to the japonica whole genome estimation in our global collection ( π = 1 . 45e-3 ) . This result suggested that the low TAC3 nucleotide diversity observed in japonica cannot be explained by a population bottleneck alone and indicated that TAC3 was strongly selected probably due to its function in controlling tiller angle during the domestication and improvement of japonica . Accordingly , GWAS , a mutant analysis and nucleotide diversity analysis demonstrated that D2 regulates the tiller angle . Thus , TAC3 and D2 contributed to the natural variation in the tiller angle and have been subjected to selection in japonica . Interestingly , the major TAC1 haplotype in japonica , Hap3 , was carried by only a small subset of indica accessions and probably introgressed from japonica . Our study on the selection of TAC1 was consistent with the previous report [38] . qTA8a and qTA8b were approximately 400 kb away on chromosome 8 and were detected via both the LMM and LR approaches . Interestingly , a locus ( marker seq-rs3945 ) associated with tiller angle at tillering stage in indica rice was fallen into the region of qTA8a and several candidate genes on this QTL were predicted by expression analysis [30] . Thus , the 400-kb region containing qTA8a and qTA8b should be further studied by developing an indica bi-parental mapping population . Additionally , there were several QTLs that co-localized with previously reported QTLs , while some QTLs were novel . Hence , TAC3 , D2 , TAC1 and other unknown genes contribute to the tiller angle variation observed in the cultivars . Rice tiller angle is a major component of plant architecture that has been subject to selection from both nature and humans over a long time period . Mining of more favorable alleles of tiller angle genes is required to achieve ideal plant architecture in rice . At the single-gene level , we identified favorable haplotypes for three tiller angle genes . Hap2 ( represented by Yuexiangzhan ) and Hap3 ( represented by 9311 ) of TAC3 decrease tiller angle , as does Hap3 ( represented by Zhenshan 97 in indica and Zhonghua 11 in japonica ) of TAC1 . Hap1 ( represented by Minghui 63 and 9311 ) and Hap4 ( represented by IR72 ) of D2 also decrease tiller angle . A three-gene combination analysis of D2 , TAC3 and TAC1 in indica showed that these three genes cause wide variation in the tiller angle , ranging from 5 . 2° to 15 . 4° on average , and combinations pyramiding decreased tiller angle haplotypes result in a more compact plant architecture . These results indicated that these genes function in the regulation of tiller angle and that their effects could be additive . Therefore , optimized haplotype-combinations among these three genes could serve as targets for designed breeding . More specifically , most of the indica cultivars carried TAC1 haplotypes that increased tiller angle; thus , the TAC1 haplotype Hap3 is the first option for improving plant architecture in indica . As most japonica cultivars carried compact haplotypes for TAC1 , TAC3 and D2 , selecting for the haplotypes of other unknown genes that decrease the tiller angle would be the first option for improving the tiller angle in japonica . In summary , the tiller angle of cultivated rice is mainly controlled by genetic factors but is also affected , to some extent , by interactions between genotype and environment . The genetic basis of tiller angle is diverse between indica and japonica rice . TAC3 , D2 and TAC1 were found to be the main factors regulating tiller angle . Introgression between two subpopulations would be an efficient means of optimizing the plant architecture through designed molecular breeding .
An association panel consisting of 529 O . sativa landraces and elite accessions was sown at the experimental farm of Huazhong Agricultural University in the winter of 2013 in Hainan and in the 2014 rice growing season in Wuhan ( China ) . The 2-year field experiment was designed with 2 replicates per year . Seven 25-day-old seedlings from these accessions were transplanted in a single row with a distance of 16 . 5 cm between plants and 26 . 4 cm between rows on December 30 , 2013 and May 12 , 2014 . The 5 plants in the middle were used to investigate the tiller angle at the heading stage . A protractor was employed to measure the angle between the most distant tillers on the two sides of the culm base , and half of the angle was treated as the tiller angle of the individual plant . The average tiller angle across 2 replicates within one year was used for GWAS . The SNPs of the 529 O . sativa accessions are available in the RiceVarMap ( http://ricevarmap . ncpgr . cn/ ) [39] . The tiller angles of the 529 O . sativa accessions are listed in S4 Table . Two-way analyses of variance were separately used to test significant difference between environments and genotypes for the whole population and two subpopulations . The analysis was run in the program Statistica 7 . 0 ( StatSoft . Tulsa , OK , USA ) . Broad-sense heritability ( H2 ) of rice tiller angle in the whole population was calculated based on the experiments using the formula: H2=δg2/ ( δg2+δge2/n+δe2/nr ) , where δg2 , δe2 and δge2 were the estimates of genetic , genotype by environment and error variances derived from the mean square expectations of two-way analysis of variance ( ANOVA ) , respectively; n was the number of environments and r was the number of replicates . Linkage disequilibrium ( LD ) was investigated based on standardized disequilibrium coefficients ( D’ ) , and squared allele-frequency correlations ( r2 ) for pairs of SNP loci were determined using the TASSEL5 . 0 program . The extent of genome-wide and chromosome-wide LD were recently reported [40] , and the average distances of LD decay at the genome-wide level in all , indica and japonica populations were 167 kb , 93 kb and 171 kb , respectively . The distances of LD decay in the regions surrounding lead SNPs identified in this study were calculated as below: First , r2 values were calculated between lead SNP and all SNPs in its upstream and downstream 2 Mb regions . Then averaged r2 of the top ten percent of r2 values in the region from 1 . 5 Mb to 2 Mb away from lead SNP were taken as background r2 . Finally the LD region was defined a continue region where r2 was 0 . 2 larger than background r2 . LD plots were generated with Haploview4 . 2 , and LD is indicated using r2 values between pairs of SNPs multiplied by 100; white , r2 = 0; shades of gray , 0<r2<1; black , r2 = 1 [41 , 42] . The SNPs of targeted genes in the 529 O . sativa samples were obtained from the RiceVarMap ( http://ricevarmap . ncpgr . cn/ ) using the gene ID [39] . The haplotypes were classified based on all SNPs with an MAF ≥ 0 . 05 in a target gene . The haplotypes contains at least 10 investigated accessions were used for comparative analysis . Duncan’s test was employed to compare the differences in the tiller angle among haplotypes using the SSPE program [43] . The whole population was previously demonstrated to present a distinct population structure [31] . The indica and japonica subpopulations and the whole population were subjected to GWAS separately because they presented sample sizes of greater than one hundred . A total of 3 , 916 , 415 , 2 , 767 , 159 and 1 , 857 , 845 SNPs ( minor allele frequency ( MAF ) ≥ 0 . 05; the number of accessions with minor alleles ≥ 6 ) were employed for GWAS using the linear mixture method ( LMM ) and linear regression ( LR ) method in the FaST-LMM program [44] . The population structure of Q matrix and kinship ( K matrix ) was taken into account as cofactor when performing association mapping using the LMM method . The effective numbers of independent SNPs and suggestive thresholds were calculated using a method described by Li et al . [45] , and 757 , 578 , 571 , 843 and 245 , 348 effective independent SNPs were found in the whole population and the indica and japonica subpopulations , respectively . The suggestive p values used as thresholds for the significance of association signals that were commonly detected in both environments by LMM were 1 . 3×10−6 for the whole population , 1 . 8×10−6 for indica and 4 . 1×10−6 for japonica . However , for significant association signals that were only detected in one environment , we utilized the more stringent p value of 6 . 0×10−7 as the threshold . A suggestive p value of 1 . 0×10−8 was employed as the threshold for the significance of association signals detected by LR , but only the top 5 loci detected by LR in each environment are presented in the results . The loci that were commonly identified by LR in both environments , but were not in the top 5 are also presented . For loci that were commonly detected by two methods , only the results of LMM are presented . To obtain independent association signals , multiple SNPs exceeding the threshold in a 5-Mb region were clustered based on an r2 of LD ≥ 0 . 25; the SNPs showing the minimum p value in a cluster were considered to be lead SNPs . The whole genomic DNA sequences of the 529 cultivar accessions were genotyped with approximately 2 . 5× coverage , and the genome was sequenced using a bar-coded multiplex sequencing approach on an Illumina Genome Analyzer II [31] . We obtained the genome sequences from RiceVarMap ( http://ricevarmap . ncpgr . cn/ ) [39] . A total of 446 O . rufipogon accessions were used to calculate nucleotide diversity ( π ) . The details of these accessions and their sequencing data have been previously reported [36] . π values were estimated at the whole-genome level , the single gene level , and the 100-kb flanking region level using SAMtools [46] . The d2-3 ( 04Z11MY27 ) , tac3-1D ( 05Z11AZ62 ) , tac3-2D ( 1B-24636 ) and 4A-02006 mutants were obtained from rice T-DNA insertion libraries from the ZH11 variety [47 , 48] and DJ variety [49 , 50] . We identified the genotypes of the mutants via PCR using the genomic primers L and R and the vector primer N ( S4 Table ) . PCR was conducted with an initial incubation step at 95°C for 5 min; a second step of 35 cycles at 95°C for 30 s , 58°C for 30 s , and 72°C for 1 min; and a final extension at 72°C for 7 min . Total RNA was extracted from different plant tissues with an RNA extraction kit using TRIzol reagent ( Invitrogen ) for quantitative real-time reverse transcription-polymerase chain reaction ( qRT-PCR ) . The total RNA ( 4g ) was reverse-transcribed using M-MLV reverse transcriptase ( Invitrogen ) . qRT-PCR was carried out in a total volume of 10 μl containing 2 . 5 μl of the reverse-transcribed products , 0 . 25 μM gene-specific primers and 5 μl of Fast Start Universal SYBR Green Master ( Rox ) superMIX ( Roche , Mannheim , Germany ) in a QuantStudio ( TM ) 6 Flex System , according to the manufacturer’s introductions . Measurements were obtained using the relative quantification method . Expression levels were normalized against expression of an ubiquitin ( UBQ ) gene . Error bars indicate standard deviations ( n = 3 ) . All primers employed for qRT-PCR are listed in S5 Table . | Tiller angle is the key component of plant architecture that greatly affect grain yield . However , few tiller angle-related genes that can be used for improving rice plant architecture have been isolated based on natural variation . Here , we identified 7 common tiller angle-related QTLs by a genome-wide association study , including the previously reported major gene TAC1 , in two environments in the 529 diverse rice accessions and dozens of QTLs specially identified in one environment . Two QTLs were validated by mutant analysis: A novel gene TAC3 , encoding a conserved hypothetical protein and preferentially expressing in the tiller base , was the candidate gene of qTA3; d2 mutant exhibited a decreased tiller angle , in addition to its previously described abnormal phenotype . A haplotype analysis identified favorable alleles of TAC3 , D2 and TAC1 in indica , which may be used for breeding plants with an ideal architecture , while they were all subjected to selection and fixed in japonica . In conclusion , there is a diverse genetic basis for tiller angle between the two subpopulations , and it is the novel gene TAC3 , together with TAC1 and D2 that greatly controls tiller angle in rice cultivars . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genome-wide",
"association",
"studies",
"plant",
"anatomy",
"quantitative",
"trait",
"loci",
"population",
"genetics",
"cereal",
"crops",
"plant",
"science",
"rice",
"model",
"organisms",
"genome",
"analysis",
"crops",
"population",
"biology",
"plants",
"research",
"and",
"analysis",
"methods",
"genome",
"complexity",
"grasses",
"genomics",
"crop",
"science",
"leaves",
"genetic",
"loci",
"agriculture",
"haplotypes",
"plant",
"and",
"algal",
"models",
"phenotypes",
"genetics",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"evolutionary",
"biology",
"introns",
"organisms",
"human",
"genetics"
] | 2016 | A Novel Tiller Angle Gene, TAC3, together with TAC1 and D2 Largely Determine the Natural Variation of Tiller Angle in Rice Cultivars |
Arthropod-borne flaviviruses such as yellow fever ( YFV ) , Zika and dengue viruses continue to cause significant human disease globally . These viruses are transmitted by mosquitoes when a female imbibes an infected blood-meal from a viremic vertebrate host and expectorates the virus into a subsequent host . Bamaga virus ( BgV ) is a flavivirus recently discovered in Culex sitiens subgroup mosquitoes collected from Cape York Peninsula , Australia . This virus phylogenetically clusters with the YFV group , but is potentially restricted in most vertebrates . However , high levels of replication in an opossum cell line ( OK ) indicate a potential association with marsupials . To ascertain whether BgV could be horizontally transmitted by mosquitoes , the vector competence of two members of the Cx . sitiens subgroup , Cx . annulirostris and Cx . sitiens , for BgV was investigated . Eleven to thirteen days after imbibing an infectious blood-meal , infection rates were 11 . 3% and 18 . 8% for Cx . annulirostris and Cx . sitiens , respectively . Cx . annulirostris transmitted the virus at low levels ( 5 . 6% had BgV-positive saliva overall ) ; Cx . sitiens did not transmit the virus . When mosquitoes were injected intrathoracially with BgV , the infection and transmission rates were 100% and 82% , respectively , for both species . These results provided evidence for the first time that BgV can be transmitted horizontally by Cx . annulirostris , the primary vector of pathogenic zoonotic flaviviruses in Australia . We also assessed whether BgV could interfere with replication in vitro , and infection and transmission in vivo of super-infecting pathogenic Culex-associated flaviviruses . BgV significantly reduced growth of Murray Valley encephalitis and West Nile ( WNV ) viruses in vitro . While prior infection with BgV by injection did not inhibit WNV super-infection of Cx . annulirostris , significantly fewer BgV-infected mosquitoes could transmit WNV than mock-injected mosquitoes . Overall , these data contribute to our understanding of flavivirus ecology , modes of transmission by Australian mosquitoes and mechanisms for super-infection interference .
The genus Flavivirus encompasses over 70 viral species including several human and animal pathogens , such as yellow fever virus ( YFV ) , dengue viruses ( DENV ) , Zika virus ( ZIKV ) , West Nile virus ( WNV ) and Murray Valley encephalitis virus ( MVEV ) which are transmitted by mosquitoes [1 , 2] . Even though most flaviviruses can replicate in Aedes , Culex , or Anopheles cells in vitro and sometimes also in vivo , flaviviruses are thought to be either Culex- ( WNV , MVEV ) or Aedes-associated ( DENV , ZIKV , YFV ) in relation to their main vector for transmission [3–5] . Horizontal transmission of these arthropod-borne viruses ( arboviruses ) occurs when the virus is ingested by a mosquito whilst it feeds on infected blood from a vertebrate host . Post ingestion by the mosquito , the virus infects and replicates in the midgut epithelial cells [6 , 7] . The virus then disseminates from the midgut cells and typically undergoes secondary replication in other tissues , such as fat bodies or neural tissues . Finally , the virus infects the cells of the salivary glands before being released into the salivary secretion when the mosquito probes a vertebrate host during feeding [6 , 7] . Several barriers to infection within the mosquito must be overcome before transmission of an arbovirus including the midgut infection and escape barriers , and the salivary infection and escape barriers [8 , 9] . Vector-competence studies aim to determine if a mosquito species can transmit an arbovirus , by evaluating if and how well the virus can overcome the infection , dissemination and transmission barriers in those mosquitoes [10–17] . These laboratory-based studies producing vectorial capacity data are crucial to determine whether these viruses pose a threat of an epidemic transmission by local mosquito species , or simply to better understand the ecological niches in which these viruses belong . Bamaga virus ( BgV ) is a flavivirus which was recently isolated from archival samples of Cx . annulirostris mosquitoes collected in 2001 and 2004 in Cape York , Far North Queensland , Australia [18] . While BgV is phylogenetically most closely related to the Australian flavivirus Edge Hill virus and other vertebrate-infecting members of the YFV group , initial in vitro characterisation experiments indicated that BgV was not able to replicate in a range of vertebrate cell lines ( monkey , chicken , rabbit ) suggesting it may have a restricted or narrow vertebrate host range [18] . In addition , injecting the virus in mice produced no disease and only caused signs of replication-associated pathology when the highest dose of virus was injected directly into the brain of the animals [18] . Despite this attenuation , BgV is classified as a vertebrate-infecting flavivirus based on its phylogenetic position , its ability to replicate to low levels in selected vertebrate cell lines ( hamster , opossum , human ) , and its dinucleotide usage bias [18 , 19] . To determine whether the virus could be horizontally transmitted by mosquitoes , laboratory-based experiments were conducted to assess BgV infection , dissemination and transmission rates in Cx . annulirostris and the closely related Cx . sitiens . Virus co- and super-infection are defined by the simultaneous or sequential infection of cells , animals or mosquitoes by two different viruses . It has been shown for a number of vertebrate-infecting flaviviruses that the level of replication or transmission of a co- or super-infecting flavivirus could be regulated by the presence of the first , both in vitro and in vivo [20] . Examples of this phenomenon include Bagaza virus which suppressed replication of Japanese encephalitis virus ( JEV ) and WNV in Culex mosquitoes upon co- and super-infection [21]; WNV and SLEV which could inhibit replication and dissemination of one another in vivo [22]; and DENV and YFV which could suppress replication of the other in vitro [23] . Furthermore , there is a subset of flaviviruses that only infect insects , and therefore , have no vertebrate hosts , but have been thoroughly studied in recent years because of their potential for co- or super-infection interference with pathogenic vertebrate-infecting flaviviruses and their high prevalence in certain mosquito populations [24 , 25] . For instance , it has been shown that WNV replication ( in vitro and in vivo ) and transmission by Culex mosquitoes could be regulated by the presence of the insect-specific flavivirus , Palm Creek virus [26 , 27] . Such interactions are important to understand in the context of risk assessment of the likelihood of an arbovirus being transmitted by local populations of mosquitoes . To further explore the phenomenon of competitive interference , we also assessed whether the presence of BgV in mosquito cells and in live mosquitoes could interfere with the replication or transmission of Culex-associated medically significant flaviviruses .
C6/36 cells ( Ae . albopictus ) were cultured in Roswell Park Memorial Institute 1640 while Vero cells ( Cercopithecus aethiops , African green monkey , kidney epithelial cells ) were cultured in Dulbecco’s Modified Eagle’s Medium . Both cell culture media were supplemented with 2–10% fetal bovine serum ( FBS ) , 50U penicillin/mL , 50μg streptomycin/mL and 2mM L-glutamine . Archival and recent mosquito homogenates were screened for the presence of BgV using the broad-spectrum Monoclonal Antibodies to Viral RNA Intermediates in Cells ( MAVRIC ) detection system [28] . Briefly , mosquitoes were collected in the wild using CO2 baited light traps as described previously [29] . Collections were sorted and female mosquitoes identified to species or genus level , pooled and homogenised in cell culture medium using glass beads and a Tissue Lyser III ( Qiagen ) for three minutes at 30Hz or following previously published methods [30] . The homogenates were clarified by centrifugation at 12 , 000 g for five minutes and filtered through a 0 . 2/0 . 8 μm sterile filter . The filtered homogenates were then inoculated on four wells of a 96 well plate pre-seeded with C6/36 mosquito cells and incubated at 28°C for 5-7days . After incubation , the cell supernatant was harvested and stored at -80°C , the cells were fixed and tested in fixed-cell enzyme-linked immunosorbent assay ( ELISA ) as described below using anti-dsRNA monoclonal antibodies MAVRIC , or pan-flavivirus monoclonal antibody ( mAb ) 4G2 . RNA was extracted from the harvested supernatant of positive samples using the Nucleospin Viral RNA extraction kit ( Macherey Nagel ) following the manufacturer’s instruction and tested by reverse-transcription PCR ( RT-PCR ) using pan-flavivirus primers and the Superscript III and Platinum Taq One-step RT-PCR kit ( Invitrogen ) following the manufacturer’s instructions [31] . Fixed cells were blocked for 30 minutes at room temperature ( RT ) in blocking buffer ( 0 . 05 M Tris/HCl ( pH 8 . 0 ) , 1 mM EDTA , 0 . 15 M NaCl , 0 . 05% ( v/v ) Tween-20 , 0 . 2% w/v casein ) . Primary mAb , at the optimal dilution in blocking buffer , was added to each well after removing the blocking buffer and incubated at 37°C for one hour . Plates were washed with PBS containing 0 . 05% Tween-20 ( PBS-T ) four times and secondary horse radish peroxidase-conjugated antibody ( goat anti-mouse , Dako ) was added diluted 1/3000 in blocking buffer and incubated at 37°C for one hour . Plates were washed six times with PBS-T and ABTS based substrate ( 1mM 2 , 2'-azino-bis ( 3-ethylbenzothiazoline-6-sulphonic acid ) with 3mM hydrogen peroxide in a 0 . 1M citrate / 0 . 2M Na2PO4 buffer pH 4 . 2 ) was added and left to develop in the dark at RT for one hour . Finally , the absorbance of each well was measured by an automated 96-well spectrophotometer at 405 nm . Positive wells were identified with a threshold of optical density higher than twice the average of mock infected wells . The virus strains used were BgV prototype CY4270 ( stock with passage number 6 , passaged only in C6/36 cells ) [18] , WNV New South Wales 2011 strain ( passaged on C6/36 , Vero and C6/36 cells successively ) [32] , MVEV strain 1–51 [33] , and Ross River virus ( RRV ) strain T-48 [34] . Titrated samples were serially diluted eight times 10-fold on C6/36 or Vero cells in 96 well plates , with four to ten replicate wells per dilution . The plates were incubated for five days at 28°C or 37°C respectively , fixed in 20% acetone , 0 . 02% bovine serum albumin in PBS , and assessed by fixed-cell ELISA as described above . C6/36 cells were incubated in suspension , rocking at RT for two hours either in mock cell culture medium or medium containing BgV to obtain a multiplicity of infection of 10 and seeded in a T175 flask to grow at 28°C for five days . After this incubation , cells were reseeded at 5x104 cells per well in 24 well plates in triplicates for each time point and virus , with one extra well seeded onto a glass coverslip , and incubated for two days at 28°C . The mock and BgV coverslips were fixed in ice cold acetone and immunolabeled by immunofluorescence assay using vertebrate-infecting flavivirus E cross-reactive mAb 4G2 , as described previously , to confirm BgV infected all the cells [18] . Cells were washed with sterile PBS once and mock- and BgV-infected cells were inoculated with either MVEV , WNV or RRV at a multiplicity of infection of 0 . 01 . After 20 minutes rocking at RT and one hour incubation at 28°C , inoculum was removed , cells washed with sterile PBS thrice and topped up with 750μL of growth medium . Culture supernatants were harvested at 1h , 8h , 24h , 48h and 72h post-infection and titrated by TCID50 on Vero cells as described above . After incubation for five days at 37°C , the cells were fixed and analysed by fixed-cell ELISA as described above , using anti-flavivirus non-structural protein 1 mAb 4G4 for MVEV and WNV ( non-reactive to BgV ) , and anti-RRV mAb G8 . Mosquitoes were collected from near Cairns ( 16o49’S , 145o42’E ) , north Queensland in April 2016 using CO2-baited passive traps and shipped to the insectary at Forensic and Scientific Services , Department of Health , Queensland Government , Brisbane , Australia [35] . Insectary conditions were 26°C and 12:12 light:dark whilst all mosquito adults were provided 15% honey water as a nutrient source . To stimulate egg production , mosquitoes were offered defibrinated sheep’s blood as a blood-meal for two hours with a Hemotek feeding apparatus ( Discovery Workshops , Accrington , Lancashire , United Kingdom ) and pig’s intestine as membrane . The blood engorged females were sorted by species and placed in 30x30x30 cm cages ( BugDorm , MegaView Science Co . , Ltd , Taiwan ) . Mosquitoes were offered blood-meals an additional eight times over 14 days . A polyethylene container containing double distilled water was added to each cage for oviposition . Egg rafts were removed daily , and first and second instar larvae fed a slurry of Tropical Fish flakes ( Wardley’s Tropical Fish Food Flakes , The Hartz Mountain Corporation , New Jersey ) , whilst third and fourth instar larvae were fed on cichlid pellets ( Kyorin Co . Ltd , Himeji , Japan ) . Pupae were removed daily and placed in cages for emergence . Ten to fifteen day old female mosquitoes were used for the BgV vector competence assessment . Mosquitoes for the virus interference experiments were collected from the suburbs of Oxley ( 27o33’S , 152o58’E ) and Banyo ( 27o22’S , 153o04’E ) in Brisbane , using CO2-baited light traps . Mosquitoes were transported to the Forensic and Scientific Services insectary and Cx . annulirostris removed , placed in a cage and used in the experiments within 24 hours of collection . Mosquitoes were exposed to BgV via feeding on a blood/virus mixture with 107 TCID50 IU/mL of virus using a Hemotek feeding apparatus . Mosquitoes were also exposed to virus by intrathoracic inoculation of approximately 220nL of BgV at a titre of 105 TCID50 IU/mL i . e . approximately 22 TCID50 IU/mosquito or 3% FBS cell culture media as mock inoculum using a Nanoject II ( Drummond Scientific , Broomall , PA ) micro injector . Mosquitoes were maintained at 28°C , high humidity and 12:12 dark:light cycle in an environmental growth cabinet ( Sanyo Electric , Gunma , Japan ) , and provided 15% honey water as a nutrient source . To assess infection , dissemination and transmission rates , mosquitoes were harvested after incubation for 8–13 days post exposure . Unfortunately , Cx . sitiens mosquitoes displayed a high mortality rate post-emergence and post-exposure , limiting the numbers available for assessment . A forced salivation method was used to assess transmission potential [36] . Briefly , legs+wings were removed from each mosquito , whose proboscis was then placed in a capillary tube with growth media with 20% FBS for two hours . The saliva samples were dispensed into 600μL of 3% FBS growth media . Bodies and legs+wings were placed separately in 2mL U-bottom tubes containing 1mL of 3% FBS growth media , in order to assess separately for virus infection and dissemination . All samples were stored at -80°C . In order to examine potential tissue tropism of BgV in mosquitoes , Cx . annulirostris exposed to virus by infectious blood-meal ( n = 25 ) or via intrathoracic inoculation ( n = 25 ) were fixed in 4% formaldehyde , 0 . 05% Triton X-100 ( BioRad ) in PBS for 24h before legs+ wings were removed and bodies transferred to 70% ethanol for storage . Mosquitoes were CO2 anaesthetized , immobilised on a refrigerated table , and injected with approximately 220nL of BgV stock virus diluted in growth medium with 3% FBS to provide a final titre of approximately 105 TCID50 IU/mL i . e . approximately 22 TCID50 IU/mosquito . Control mosquitoes were injected with growth medium only . After 7–8 days incubation at 28°C , 12:12 light:dark cycle and high relative humidity , mosquitoes were offered a blood-meal containing 106 TCID50 IU/mL of WNV . Mosquitoes were again incubated at 28°C , 12:12 dark:light cycle and high relative humidity , before bodies , legs+wings and saliva expectorates were collected seven or ten days post-blood-meal as described above , and stored at -80°C . Bodies and legs+wings samples from the vector competence study were homogenised with a metal bead in a Tissue Lyser III ( Qiagen ) for three minutes at 30Hz , clarified by centrifugation at 12 , 000 g for five minutes and filtered through a 0 . 22μm sterile filter . Mosquito body homogenates were titrated by TCID50 on C6/36 cells as described above . Undiluted supernatant from homogenised legs+wings of positive mosquito bodies were directly inoculated on C6/36 cells , in four wells of a 96 well plate . Finally , saliva expectorates from mosquitoes positive for dissemination ( virus detected in legs+wings ) were titrated to determine a virus titre in the saliva . Replication was assessed by fixed-cell ELISA ( see above ) with pan-flavivirus mAb 4G2 [37] . Additionally , 31 BgV injected Cx . annulirostris mosquitoes were harvested as whole mosquitoes , to be included in the overall infection rate , and homogenised as described above . These were tested for presence of BgV by directly inoculating homogenate on C6/36 cells as described above for the legs+wings and performing a fixed-cell ELISA with mAb 4G2 . Bodies of BgV-injected and mock-injected mosquitoes from the virus interference experiments were processed similarly as above , titrated on Vero cells and analysed by fixed-cell ELISA with mAb 4G4 ( WNV reactive and BgV non-reactive ) . Vero cells were used here to prevent BgV from interfering with WNV replication further , since these vertebrate cells do not support BgV replication . To assess the infection status of the BgV-injected mosquitoes , these homogenates were also inoculated in four wells of a 96 well plate of C6/36 cells and analysed by fixed-cell ELISA with mAb 1B7 , which is BgV reactive and WNV non-reactive . The samples were not titrated on C6/36 cells as the potential presence of WNV in the homogenates could have interfered with BgV and altered the titres obtained . Legs+wings from WNV positive bodies were inoculated in four wells of 96 well plates pre-seeded with C6/36 cells , and Vero cells incubated at 28 or 37°C , respectively , for six days before being analysed by fixed-cell ELISA with either mAb 4G4 ( Vero ) or 1B7 ( C6/36 ) to test for presence of either WNV or BgV , respectively . Similar to body homogenates , saliva from mosquitoes with positive bodies were titrated on Vero cells and inoculated on C6/36 cells as described above and analysed by fixed-cell ELISA with either mAb 4G4 or 1B7 . Fixed mosquitoes were paraffin-embedded prior to immunohistochemistry ( IHC ) as per routine processing described previously [19] . Five μm sections , collected on charged slides , were immuno-labelled for BgV using a cocktail of BgV reactive mAbs ( e . g . 4G2 [38] , 6B6C [39] , 1D1 [40] and 1B7 [41] ) or BgV-specific mouse serum [18] following a previously described protocol [26] . The mAb hybridoma supernatants used in this protocol were tested in fixed-cell ELISA with cells fixed with 4% formaldehyde in PBS with 0 . 05% Triton X-100 to empirically determine the optimal dilutions to use on the formaldehyde fixed mosquitoes in IHC . The titres obtained by TCID50 were determined using Reed and Muench’s guidelines [42] . For the BgV vector competence experiments , the titre of virus in bodies and saliva expectorates of injected and bloodfed Cx . annulirostris and Cx . sitiens were compared using an unpaired parametric t-test . The MVEV , WNV and RRV titres in the in vitro super-infection experiment were analysed using an unpaired parametric t-test . For the in vivo virus interference experiments , Fisher’s exact tests were used to compare WNV infection , dissemination and transmission rates between BgV-infected and mock-infected Cx . annulirostris . The titres of WNV positive body samples were statistically analysed using an unpaired parametric t-test . The titres of WNV positive saliva could not be statistically analysed considering that one of the groups only had one positive sample . All analyses were conducted using Graphpad Prism Version 7 ( GraphPad Software , Inc , San Diego , USA ) .
There are only three known isolates of BgV , all detected in Cx . sitiens subgroup mosquitoes collected on Cape York Peninsula , Far North Queensland , Australia between 2001 and 2004 [18] . To further determine the prevalence of BgV in Culex and other species in other genera , 811 additional mosquito pools ( pool size ranging from 1 to 107 ) from the Aedeomyia , Aedes , Anopheles , Coquillettidia , Culex , Culiseta , Mansonia , Uranotaenia and Verrallina genera , encompassing at least 31 species , were screened for BgV ( Table 1 ) . The mosquito homogenates were inoculated onto C6/36 mosquito cells for isolation and screened using anti-dsRNA mAbs MAVRIC and/or using pan-flavivirus mAb 4G2 in ELISA and/or RT-PCR using pan-flavivirus primers [31] . No BgV isolates were recovered from this range of mosquito homogenates ( Table 1 ) . To determine whether this vertebrate-restricted virus could be horizontally transmitted by field collected mosquitoes , laboratory reared progeny of wild Cx . annulirostris and Cx . sitiens collected from Cairns , northern Queensland , were exposed to BgV using methods previously published [35] . After being exposed to an infectious blood-meal with a titre of 107 TCID50 infectious units per milliliter ( TCID50 IU/mL ) , infection rates were 11 . 3% ( 8/71 ) and 18 . 8% ( 3/16 ) for Cx . annulirostris and Cx . sitiens , respectively ( Table 2 and Table 3 ) . BgV was detected in the legs+wings of 8/8 positive blood-fed Cx . annulirostris mosquitoes whilst none of the three positive Cx . sitiens mosquitoes had virus disseminated in their legs+wings ( Table 2 and Table 3 ) . Half of the positive Cx . annulirostris had detectable levels of virus in their saliva ( 4/8 ) , resulting in an overall transmission rate of 5 . 6% ( 4/71 ) for Cx . annulirostris; there was no detectable transmission for Cx . sitiens ( Table 2 and Table 3 ) . In addition to exposing the mosquitoes to BgV via a blood-meal , Cx . annulirostris and sitiens mosquitoes were injected intrathoracically with the virus , in order to bypass the midgut infection and escape barriers , and to provide a group with a controlled virus dose . Injecting BgV intrathoracically achieved 100% infection and dissemination rates ( 45/45 for Cx . annulirostris and 11/11 for Cx . sitiens ) . The virus was found in high prevalence in the saliva of injected mosquitoes , with 82 . 2% ( 37/45 ) and 81 . 8% ( 9/11 ) transmission rates for Cx . annulirostris and Cx . sitiens respectively ( Table 2 and Table 3 ) . The level of BgV amplification in vivo was measured in injected and bloodfed mosquitoes . Whilst the mosquitoes were injected with the equivalent of approximately 20 TCID50 IU , body titres of 106 TCID50 IU per injected Cx . annulirostris were recovered on average eight to ten days post exposure , and 106 . 6 TCID50 IU per injected Cx . sitiens , which indicated successful replication of the virus in the mosquito tissues ( Fig 1 ) . There was a significant difference between the titres in the bodies of blood-fed Cx . sitiens and Cx . annulirostris , with a higher titre in Cx . annulirostris ( P = 0 . 0149 ) . This was also the case in the bodies of injected mosquitoes , with slightly higher titres on average detected in Cx . sitiens ( P = 0 . 0058 ) . Titres were significantly lower in the mosquito bodies of the blood-fed mosquitoes than in the injected mosquitoes ( P < 0 . 0001 for both species ) . In contrast , virus titres in saliva expectorates of injected mosquitoes were not significantly different between Cx . sitiens and Cx . annulirostris ( P = 0 . 3712 ) . Similarly , titres in the saliva were not significantly different between injected and blood-fed Cx . annulirostris , although this could be the result of a longer extrinsic incubation for bloodfed mosquitoes ( P = 0 . 6977 ) . BgV was detected by IHC in the midgut epithelial cells and neuronal cells of 8/25 bloodfed ( harvested 14 days post-infection ) and 20/25 injected Cx . annulirostris ( harvested 10 days post-infection ) . There was more virus antigen present in the midgut of blood-fed mosquitoes , while injected mosquitoes displayed more antigen in the neuronal cells ( Fig 2 ) . Viral antigen was also detected in the fat bodies proximal to the gonads and in the salivary glands . To determine whether prior infection with BgV affects replication of pathogenic flaviviruses , mosquito C6/36 cells were either mock- or BgV-infected for five days , and subsequently infected with RRV , MVEV or WNV . BgV-infected cells were shown to be significantly less permissive to WNV and MVEV super-infection than mock-infected cells ( Fig 3A ) . Indeed , no MVEV replication could be detected in BgV infected cells at any time points , while the MVEV titre increased over time in mock-infected cells . The difference was statistically significant at 48h and 72h with P < 0 . 0001 . Significantly lower titres were attained for WNV in BgV infected cells at all time points , with P = 0 . 0003 at 24h , P = 0 . 0061 at 48h and P = 0 . 0035 at 72h . This super-infection interference phenomenon was not observed for RRV ( with P = 0 . 3252 at 24h , P = 0 . 1427 at 48h and P = 0 . 4145 at 72h ) , suggesting BgV specifically interfered with flavivirus replication in vitro . To examine whether BgV interfered with infection and transmission of pathogenic flaviviruses in vivo , Cx . annulirostris mosquitoes were injected with either growth medium or BgV ( approximately 220nL at 105 TCID50 IU/mL ) . Seven to eight days later , the mosquitoes were offered a blood/virus mixture containing 106 TCID50 IU/mL of WNV . At seven days post-infection , no mosquito bodies tested positive for WNV in either group ( n = 29 mock-injected and n = 23 BgV-injected ) , while 15/23 BgV-injected mosquitoes had detectable levels of BgV . At ten days post-infection , there was no significant difference in the number of mosquito bodies positive for WNV between mock- ( 10/31 ) and BgV-infected ( 8/27 ) mosquitoes ( P > 0 . 9999 ) , and the titres were not significantly different ( P = 0 . 6019 ) ( Table 4 , Fig 3B ) . Similarly , there was no significant difference in WNV dissemination rates between the two groups ( 9/10 mock-injected and 8/8 BgV-injected ) ( P > 0 . 9999 ) ( Table 4 ) . However , fewer mosquitoes had detectable amounts of WNV in their saliva in the BgV-infected group ( 1/8 ) compared to the mock-infected group ( 6/10 ) ( Table 4 ) , although the difference was not significant ( P = 0 . 0656 ) . The single BgV-infected mosquito with WNV detected in saliva had a WNV saliva titre at the limit of detection , ( 101 . 97 TCID50 IU/mL ) , which was lower than the average for WNV-positive saliva of mock-injected mosquitoes ( 104 . 42 TCID50 IU/mL ) ( Fig 3B ) .
The results from experiments performed in Culex mosquitoes provide evidence that BgV is likely transmitted and maintained in the environment using a classical arbovirus transmission cycle . This cycle comprises ingestion of an infectious blood-meal , replication in various tissues and transmission via the saliva of the infected mosquito . The localization of BgV in various mosquito tissues was consistent with the classical model for flavivirus dissemination in mosquitoes but in contrast to what we have previously reported for Australian insect-specific flaviviruses , which appear to be restricted to the midgut of their mosquito hosts [19 , 26] . Despite this confirmation that BgV could be part of a classical arbovirus transmission cycle , the relatively low proportion of Cx . annulirostris mosquitoes which could transmit the virus ( 5 . 6% ) suggested that Culex mosquitoes were not highly competent vectors of BgV as they are of other vertebrate-infecting flaviviruses such as WNV or JEV ( > 50% transmission rate for both ) in Australia [3 , 17 , 43] . Indeed , Cx . annulirostris and Cx . sitiens appeared to express a midgut infection barrier , with less than 25–30% of mosquitoes infected following ingestion of an infectious blood-meal of BgV . Cx . annulirostris did not appear to express a midgut escape barrier , as all infected mosquitoes had a disseminated infection and 50% of these transmitted the virus . Although numbers were relatively low , none of the infected Cx . sitiens had a disseminated infection , suggesting the presence of a midgut escape barrier in this species . Further evidence for the presence of these various barriers was provided by the results of the IT inoculations , whereby all mosquitoes from both species possessed a disseminated infection and the majority of these transmitted the virus . The viral titres recovered from the injected mosquito samples ( bodies and saliva ) were similar to what has been found for WNV-injected Culex mosquitoes [26] , while the titres in the blood-fed mosquito samples ( bodies and saliva ) were lower than for WNV in Culex mosquitoes [26 , 43] . We conducted our in vivo experiments with Cx . annulirostris and Cx . sitiens , as these species belong to the Cx . sitiens group , which are the only taxonomic group that have yielded detections of BgV . It is possible that members of other mosquito genera may be more efficient vectors of BgV and may play a greater role in maintaining the virus in nature . Indeed , the infection and transmission rates of Culex-associated flaviviruses such as WNV or JEV are much lower in Aedes mosquitoes ( <35% infection; <15% transmission and 27% infection; 25% transmission respectively ) than in Culex ( >70% infection; >50% transmission and >90% infection; >50% transmission respectively ) , and more similar to what was observed for BgV in Culex [3 , 17] . Additionally , the virus may have been present in other genera during the time these collections were undertaken , but remained undetected . The preponderance for BgV to be detected only in Cx . sitiens subgroup mosquitoes could be an artefact of the emphasis placed on processing Culex spp . which are the primary vectors of JEV , during investigations of this virus in northern Australia [29] . Other genera were discarded and so were underrepresented in the original study that yielded BgV . Further evidence that BgV may be transmitted by other mosquito species is provided by the phylogenetic position of BgV in the YFV group , in which the viruses are thought to be Aedes-associated rather than Culex-associated . Finally , in immuno-assays with anti-dsRNA mAbs , BgV can only be detected in cells that have been fixed in 100% acetone , as opposed to our standard fixative buffer with 20% acetone for ELISA , which has previously been demonstrated for DENV serotypes 1 and 2 , indicating that BgV may share a similar mode of replication with these two Aedes-associated flaviviruses in vitro [28] . Collectively , this suggests that even though BgV can be horizontally transmitted by Culex mosquitoes and has only ever been isolated from Culex mosquitoes , it might be preferentially Aedes-associated rather than Culex-associated . Clearly , further in vivo experiments with other genera , particularly Aedes spp . , are needed to incriminate other mosquito species that could serve as vectors of BgV . We attempted to address the underrepresentation of other mosquito genera by screening an additional 811 archival and recent pools from a wide range of species , collected from a broad geographical range in Australia and Papua New Guinea . No BgV isolates could be recovered from these pools , despite the tested mosquitoes being from a wide geographical ( Australian states of New South Wales , Queensland , Northern Territory , Western Australia , as well as Papua New Guinea ) and temporal ( 1973 to 2018 ) distribution . The low prevalence observed for BgV in Australasian mosquitoes is not unusual for a vertebrate-infecting flavivirus [44 , 45] , with many thousands of mosquitoes often yielding a single positive sample only . This low prevalence is reflective of the intricate virus transmission cycle between specific amplifying vertebrate hosts and competent mosquito vectors . Such a low number of positive samples , found only in one mosquito species and one location , Cx . annulirostris from Cape York Peninsula , indicates a small ecological niche for this virus . It is possible that BgV has a cryptic vertebrate host found only in certain parts of Australia , which would also fit with its apparent host-restriction in some vertebrate cells [18] . There could also be a discrepancy between the optimal amplifying vertebrate host of BgV and the feeding preferences of its optimal mosquito vector , resulting in low incidence rates in Australasian mosquitoes . Indeed , the most competent mosquito vector species may not include the best BgV amplifying vertebrate host in its blood feeding patterns . This phenomenon has been shown to potentially reduce JEV transmission in certain areas , as Cx . annulirostris preferentially feeds on cattle and wallabies , while the virus is not amplified to sufficient levels in these animals to be transmitted [45 , 46] . It should however be noted that the sample size for each mosquito population tested here was relatively small . Thus , further assessment of the prevalence of BgV in mosquito species collected in the Cape York region should be undertaken , along with assessment of the prevalence of BgV-specific antibodies in major vertebrate species in the area , such as agile wallabies , in order to confirm the primary vector and vertebrate host of BgV . These considerations will help better understand its ecology , host-restriction , evolution and potential to emerge as a virus capable of causing disease in humans or other animals . Considering that flaviviruses have previously been shown to interfere with the replication of related viruses in vitro and in vivo , and that BgV was isolated in a cohort of samples that also yielded several other flavivirus isolates , super-infection interference studies were performed [18 , 29] . In our laboratory setting , it was clear that BgV could interfere with the replication and transmission of pathogenic flaviviruses both in C6/36 Ae . albopictus cells in vitro and in Culex mosquitoes in vivo . The data generated here demonstrated that primary infection with BgV completely prevented replication of MVEV in C6/36 cells , strongly suppressed replication of WNV , but had no significant effect on the alphavirus RRV , suggesting the interference was specific to flaviviruses . Additionally , it was shown that while BgV did not seem to interfere with WNV replication or dissemination in Culex mosquitoes , WNV transmission was inhibited in the presence of BgV . Indeed , there was no difference in the number of mosquitoes positive for WNV or in the WNV titres in the bodies of mock- and BgV-injected WNV-blood-fed mosquitoes . These results suggested that for BgV , the midgut was not the main site for interference . However , it was clear that BgV interfered with WNV after the virus had escaped the midgut barrier , since the number of WNV-positive mosquito saliva and the WNV titre in these saliva samples were lower in BgV-injected mosquitoes than in mock-injected mosquitoes . These data were similar to what was found with lineage II insect-specific flavivirus , Nhumirim virus , interfering with WNV in Cx . quinquefasciatus mosquitoes [47] , but differ from what was found for Australian lineage I insect-specific flavivirus Palm Creek virus , which seemed to interfere with WNV replication in the midgut epithelial cells of infected mosquitoes [26] . However , it should be noted that in each study , intrathoracic inoculation was used to infect the mosquitoes with the primary virus , in this case BgV . While it was shown that BgV was present in detectable levels in the midgut epithelial cells of injected mosquitoes ( Fig 2 ) , more BgV was present in the midgut of mosquitoes orally infected with BgV . This is consistent with the midgut epithelium being the first cells encountered by a virus in an infectious blood-meal . Thus , mosquitoes orally infected with BgV , presumably the natural route of infection , may provide more resistance to subsequent infection of the midgut . In most instances and in accordance with our in vitro data , reports suggest that flaviviruses do not interfere with the replication of viruses from other families or genera such as alphaviruses , parvoviruses , or baculoviruses [48–54] . However , other authors have reported that an interfering interaction between flavivirus and other virus families such as rhabdoviruses , parvoviruses or alphaviruses can occur [55–58] . These discrepancies are most likely due to experimental design differences , such as the virus species used , the order of infections or whether the experiment is a co-infection or a super-infection . Even though these interactions can happen in laboratory models , in vitro or in vivo , they may not actually occur in nature , since the prevalence of arthropod-borne viruses in mosquito populations is quite low , as mentioned above . In conclusion , we have successfully shown that the flavivirus BgV can be transmitted horizontally by Cx . annulirostris , a member of the Cx . sitiens subgroup , from which all BgV isolates have been obtained , and thereby be included in a classical arbovirus transmission cycle . We have also demonstrated that BgV can interfere with Culex-associated WNV transmission in vivo . Further investigations could include vector competence studies of Aedes mosquitoes for BgV to draw comparisons with the data presented here , as well as super-infection interference experiments with Aedes-associated flaviviruses . Future work should also comprise studying the other side of the transmission cycle: the host-restriction of BgV in vertebrates . Overall , the presented data contribute to elucidating the ecology of an Australian flavivirus and help to further the understanding of mechanisms of interference between flaviviruses in mosquitoes . | Mosquito-borne flaviviruses include medically significant members such as the dengue viruses , yellow fever virus and Zika virus . These viruses regularly cause outbreaks globally , notably in tropical regions . The ability of mosquitoes to transmit these viruses to vertebrate hosts plays a major role in determining the scale of these outbreaks . It is essential to assess the risk of emergence of flaviviruses in a given region by investigating the vector competence of local mosquitoes for these viruses . Bamaga virus was recently discovered in Australia in Culex mosquitoes and shown to be related to yellow fever virus . In this article , we investigated the potential for Bamaga virus to emerge as an arthropod-borne viral pathogen by assessing the vector competence of Cx . annulirostris and Cx . sitiens mosquitoes for this virus . We showed that Bamaga virus could be detected in the saliva of Cx . annulirostris after an infectious blood-meal , demonstrating that the virus could be horizontally transmitted . In addition , we showed that Bamaga virus could interfere with the replication in vitro and transmission in vivo of the pathogenic flavivirus West Nile virus . These data provide further insight on how interactions between viruses in their vector can influence the efficiency of pathogen transmission . | [
"Abstract",
"Introduction",
"Material",
"and",
"methods",
"Results",
"Discussion"
] | [
"invertebrates",
"dengue",
"virus",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"microbiology",
"vertebrates",
"saliva",
"animals",
"viruses",
"rna",
"viruses",
"insect",
"vectors",
"infectious",
"diseases",
"medical",
"microbiology",
"microbial",
"pathogens",
"arboviral",
"infections",
"viral",
"replication",
"disease",
"vectors",
"insects",
"arthropoda",
"mosquitoes",
"eukaryota",
"west",
"nile",
"virus",
"anatomy",
"flaviviruses",
"virology",
"viral",
"pathogens",
"physiology",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"viral",
"diseases",
"organisms"
] | 2018 | The recently identified flavivirus Bamaga virus is transmitted horizontally by Culex mosquitoes and interferes with West Nile virus replication in vitro and transmission in vivo |
The human genitourinary tract is a common anatomical niche for polymicrobial infection and a leading site for the development of bacteremia and sepsis . Most uncomplicated , community-acquired urinary tract infections ( UTI ) are caused by Escherichia coli , while another bacterium , Proteus mirabilis , is more often associated with complicated UTI . Here , we report that uropathogenic E . coli and P . mirabilis have divergent requirements for specific central pathways in vivo despite colonizing and occupying the same host environment . Using mutants of specific central metabolism enzymes , we determined glycolysis mutants lacking pgi , tpiA , pfkA , or pykA all have fitness defects in vivo for P . mirabilis but do not affect colonization of E . coli during UTI . Similarly , the oxidative pentose phosphate pathway is required only for P . mirabilis in vivo . In contrast , gluconeogenesis is required only for E . coli fitness in vivo . The remarkable difference in central pathway utilization between E . coli and P . mirabilis during experimental UTI was also observed for TCA cycle mutants in sdhB , fumC , and frdA . The distinct in vivo requirements between these pathogens suggest E . coli and P . mirabilis are not direct competitors within host urinary tract nutritional niche . In support of this , we found that co-infection with E . coli and P . mirabilis wild-type strains enhanced bacterial colonization and persistence of both pathogens during UTI . Our results reveal that complementary utilization of central carbon metabolism facilitates polymicrobial disease and suggests microbial activity in vivo alters the host urinary tract nutritional niche .
The recent revival of interest in the relationship between bacterial metabolism and host-pathogen interactions has deepened our understanding of pathogen colonization and growth in vivo [1] , [2] , [3] , [4] . Consequently , central metabolism must be considered essential to virulence because bacterial pathogens must use nutrients available within the host niche to cause disease [5] , [6] . The relationship between the pathogen's available carbon and energy sources , or host nutritional niche [7] , and pathways required for replication in vivo has been demonstrated for a variety of pathogenic microbes . Extraintestinal pathogenic E . coli require peptide import systems , the TCA cycle , and gluconeogenesis to consume amino acids and peptides in the urinary tract [8] , while intestinal pathogenic E . coli require pathways to catabolize multiple sugars available in the intestine [9] . Salmonella enterica serovar Typhimurium exploits the inflammatory response of the host that creates an alternative electron acceptor to allow the pathogen to respire and compete with anaerobic gut residents [10] . The energetic consequences of modulating respiratory chain components and proton motive force can also promote pathogen survival in the face of bactericidal activities of the host [11] . The contribution of central carbon pathways to pathogenesis has been shown for numerous intracellular and extracellular nutritional niches occupied by pathogens . Expression-based and genetic approaches using the model cytosolic pathogen , Listeria monocytogenes , indicate that gluconeogenesis and uptake and catabolism of glycerol and dihydoxyacetone are required for bacterial replication in vivo [12] , [13] , [14] , [15] . These findings are supported by the observation that disruption of glucose uptake has no effect on L . monocytogenes intracellular replication [16] , suggesting that glycerol is the preferred carbon source for Listeria in vivo . Shigella flexneri , which also replicates in the host cell cytosol , requires glycerol-3-phosphate as a carbon source [17] . Interestingly , enteroinvasive E . coli ( EIEC ) , which are genetically-related to S . flexneri , also use 3-carbon substrates as a carbon source in vivo . Because chorismate , GMP , and thymidylate synthesis have been found essential for S . flexneri replication in vivo [18] , [19] and de novo synthesis of amino acids is extensive , it is believed that glycerol metabolism and anabolic pathways in this bacterium and EIEC may be important for bacteria that replicate within epithelial cells during intestinal infection . Mycobacterium tuberculosis that replicates intracellularly within phagocytes , utilizes fatty acids in addition to glycerol or glycerol-3-phosphate as a carbon source in macrophages [20] . Seminal studies show that in vivo carbon metabolism in M . tuberculosis is dependent on fatty acid catabolism and the glyoxylate shunt through the TCA cycle [21] , [22] , [23] . The collective requirement for glycerol catabolism for L . monocytogenes , S . flexneri , EIEC , and M . tuberculosis growth in vivo likely indicates that glycerol is a readily available carbon source inside host cells and extends Rolf Freter's nutrient-niche hypothesis [7] to include available nutrients within a eukaryotic cell . However , despite occupying a similar host microenvironment , intracellular glycerol is not the preferred carbon source for S . enterica serovar Typhimurium because glucose import , glycolysis , and the oxidative TCA cycle are required for Salmonella to colonize the intestine and replicate within host phagocytes [24] , [25] , [26] , [27] . This key difference in preferred carbon source in vivo could reflect a Salmonella fitness adaptation for facing increased competition with a diversity of luminal gut anaerobes . Extracellular or luminal colonizers , including both commensal and pathogenic E . coli , are able to occupy the host gastrointestinal tract , yet their nutritional requirements for carbon metabolism in vivo have key differences . Colonization studies using the streptomycin-treated mouse model in combination with transcriptional profiling during culture in mucus demonstrated that the ED pathway , and gluconate or other sugar acids , are required for intestinal growth of commensal E . coli [28] . EHEC requires similar central metabolic pathways as commensal strains , however , EHEC colonization in vivo requires the catabolism of up to six additional sugars [9] . EHEC also utilizes glycolytic substrates and switches to gluconeogenic substrates when present in the intestine with commensal E . coli , which solely utilizes glycolytic pathways for in vivo growth [29] . This finding , that competition in vivo can alter preferred routes of carbon flux through the central pathways , introduces the notion that studying polymicrobial interactions during host colonization is essential to understand the relationship between bacterial metabolism and pathogenesis . For extraintestinal pathogenic E . coli ( ExPEC ) , it has been shown that D-serine metabolism and acetogenic growth are important during colonization of the urinary tract [30] , [31] . In previous work , we have demonstrated that the import of peptides , gluconeogenesis , and the TCA cycle are required for E . coli during extraintestinal infection , while glycolysis and the pentose phosphate pathway are dispensable [8] . This indicates E . coli has to synthesize sugars from amino acids ( gluconeogenesis ) while enzymes for sugar catabolism have no affect on fitness . Although less is known about the in vivo metabolism of Proteus mirabilis , another important urinary tract pathogen , it would be expected to have the same enzymatic requirements during infection . Attenuated strains of P . mirabilis have been identified with mutations in genes that encode proteins involved in gluconate and pyruvate metabolism , and in enzymes of the TCA cycle , using signature-tagged mutagenesis [32] , [33] . These earlier studies are supported by a recent comparison of global gene expression studies from E . coli [34] , [35] and P . mirabilis [36] that indicated many similarities and some subtle differences may exist in vivo between these uropathogens during experimental infection of the urinary tract . To better understand the relationship between the host nutritional niche and pathogen growth , we used defined mutants , each defective in specific metabolic pathways , to directly examine the in vivo metabolism for two bacterial pathogens that occupy the same host niche and likely have access to the same nutrients during infection . Unexpectedly , we found remarkably divergent in vivo requirements for central pathways between these two pathogens during UTI by assessing the in vivo fitness of strains containing mutations in pgi , pfkA , tpiA , pykA , gnd , talB , edd , sdhB , fumC , frdA , and pckA , in both uropathogenic E . coli CFT073 and P . mirabilis HI4320 ( Fig . 1 ) . Because the urinary tract is a normally sterile environment , in this current study , we not only further characterized the role of central metabolism during host colonization for both pathogenic E . coli and P . mirabilis in mono-species infection , but it was also possible to develop a polymicrobial infection model using the host urinary tract as an unoccupied vessel . Using this new model and , as their complementary utilization of central pathways suggested , we found that co-infection with E . coli and P . mirabilis wild-type strains enhanced bacterial colonization and persistence of both pathogens during UTI . These findings help explain the molecular and biochemical basis of polymicrobial infection in the urinary tract .
Despite extensive biochemical and in vitro studies of the model organism E . coli , characterization of central carbon pathways for extraintestinal pathogenic E . coli during infection is considerably less well understood than well-considered virulence factors [6] . Recently , a uropathogenic isolate was used to investigate pathogenic E . coli central metabolism in an infection model and found that in contrast to commensal E . coli , glycolysis is dispensable for extraintestinal pathogenic E . coli during colonization , while gluconeogenesis is required during infection [8] . One limitation from that study is the glycolytic enzymes investigated in that work could also play a role in gluconeogenesis . To address this , additional glycolytic mutants were constructed in E . coli CFT073 , a prototype strain isolated from the blood and urine of a patient with acute pyelonephritis and urosepsis [37] , [38] . In addition to strains lacking tpiA ( triose phosphate isomerase ) and pgi ( phosphoglucose isomerase ) , mutants in irreversible glycolytic steps involving both 6-carbon ( pfkA; 6-phosphofructokinase transferase ) and 3-carbon ( pykA; pyruvate kinase ) substrates were constructed and tested in competitive infections with the parental E . coli CFT073 strain . Using the well-established murine model of ascending infection [39] , we found that disruption of either the preparative or substrate level phosphorylation stages of glycolysis had no effect on the ability of E . coli to compete with wild-type CFT073 during experimental infection ( Fig . 2A ) . Because the growth medium within the urinary tract , urine , is a dilute mixture of amino acids and peptides [40] , it is not unexpected that glycolysis would be dispensible during UTI . The composition of urine and its relative lack of available carbohydrates ( except under diabetic condition ) , along with the lack of a significant contribution of glycolysis for E . coli during experimental UTI , predicts that glycolysis mutants in P . mirabilis , another common urinary tract pathogen , would have no apparent fitness defect in vivo . To test this , mutations were constructed in P . mirabilis HI4320 , in the same glycolysis genes tested for E . coli: pgi , pfkA , tpiA , and pykA . Unexpectedly , any mutation that disrupted glycolysis in P . mirabilis resulted in a significant fitness defect using the same model of ascending infection ( Fig . 2B ) . With the exception of pykA that demonstrated a fitness defect in the bladder ( P = 0 . 031 ) , all of the remaining P . mirabilis glycolysis mutants tested were out-competed by wild-type parental HI4320>1 , 000-fold ( P<0 . 050 ) in both bladders and kidneys ( Fig . 2B ) . The dramatic differential requirement for glycolysis during infection between E . coli and P . mirabilis was not due to differences during in vitro growth . In both E . coli and P . mirabilis , only mutants in pfkA , pgi , and tpiA demonstrated the expected growth defect in defined medium containing glucose as the sole carbon source; growth rates of pykA mutants in defined medium containing glucose was similar to wild-type for both strains ( Fig . 2C and D ) . Both sets of glycolysis mutants also demonstrated growth rates similar to the parental strains in LB medium ( Fig . 2E and F ) . It was possible to complement the in vitro growth defect for the P . mirabilis tpiA mutant by introducing the tpiA gene from wild-type HI4320 into the mutant strain on a low copy plasmid ( Fig . 2H ) . The re-introduction of the wild-type tpiA allele also restored the ability of the mutant strain to colonize the urinary tract; in vivo complementation resulted in a complete reversal of the tpiA fitness defect during competitive infection with the parental wild-type HI4320 harboring an empty vector control plasmid ( Fig . 2I ) . It was also possible to heterologously complement the growth defect for the E . coli tpiA mutant in defined medium containing glucose by introduction of the wild-type tpiA allele cloned from P . mirabilis HI4320 ( Fig . 2G ) . Previously , it has been shown that mutation of E . coli transaldolase A gene ( talA ) does not negatively affect fitness during extraintestinal infection despite TalA being induced by CFT073 when cultured in human urine [8] . This suggests that the non-oxidative pentose phosphate pathway does not significantly contribute to pathogen fitness during urinary tract infection . To better characterize the contribution of the isomerizations of the non-oxidative pentose phosphate pathway in vivo , an additional transaldolase mutant , transaldolase B ( talB ) , was constructed in E . coli CFT073 . TalB is the major transaldolase in E . coli that transfers a three-carbon moiety from a C7 molecule to glyceraldehyde-3-P ( C3 ) to form erythrose-4-P ( C4 ) and fructose-P ( C6 ) . This stage of the pentose phosphate pathway is reversible , and thus , can be uncoupled from the oxidative decarboxylation reactions that produce NADPH . While loss of the major transaldolase , TalB , did not affect E . coli fitness during UTI ( Fig . 3A ) ; P . mirabilis talB mutant bacteria were out-competed>100-fold by wild-type in both the bladders and kidneys ( P<0 . 003 ) ( Fig . 3B ) . One possibility for a difference between E . coli and P . mirabilis requirement in the non-oxidative pentose phosphate pathway is the redundancy of transaldolase in E . coli [41] . P . mirabilis strains encode a single transaldolase enzyme ( TalB ) , while E . coli strains have both TalA and TalB , which catalyze identical reactions for the cell . To determine if the lack of a fitness defect for the E . coli talB mutant ( Fig . 3A ) , is due to functional redundancy , we tested the talA single mutant and constructed and tested a talA talB double mutant strain in competitive infections with the parental CFT073 wild-type strain . In these studies , we found that lack of talA resulted in the wild-type being outcompeted by the single mutant in the bladder ( P = 0 . 043 ) ( Fig . 3C ) . Although not as striking as the talB mutant in P . mirabilis , loss of both talA and talB in E . coli resulted in the transaldolase double mutant being out-competed by wild-type CFT073>5 . 0-fold in bladders and kidneys ( P<0 . 050 ) ( Fig . 3D ) . To distinguish the relative importance of the oxidative branch of the pentose phosphate pathway from the non-oxidative transaldolase-containing branch , a mutant in phophogluconolactonate ( gnd ) and a mutant defective in gluconate catabolism , 6-phosphoglyconate dehydrase ( edd ) , were tested in competitive infections with the parental CFT073 wild-type strain . E . coli lacking either the oxidative pentose phosphate pathway ( gnd ) , or the ED pathway ( edd ) were recovered from bladders and kidneys in numbers not significantly different from the wild-type strain ( median CI = 1 ) ( Fig . 3A ) . Surprisingly , unlike the lack of contribution for oxidative production of NADPH in the pentose phosphate pathway for E . coli , P . mirabilis mutants in gnd were out-competed>100-fold by wild-type in both the bladders and kidneys ( P<0 . 003 ) ( Fig . 3B ) . Previously , signature-tagged mutagenesis identified a P . mirabilis edd transposon insertion as attenuated during experimental UTI [33] , however , the attenuation caused by the disruption of edd was not confirmed by testing a ‘clean’ isogenic mutant strain in vivo . Despite this , it was reasonable to expect that , similar to the different requirement for glycolysis between E . coli and P . mirabilis during infection , P . mirabilis may require the capacity to metabolize gluconate via the Entner-Duodoroff pathway in vivo . Consistent with this , we found that in contrast to the findings with E . coli , the P . mirabilis edd mutant was significantly out-competed in both the bladders and kidneys by the parental HI4320 strain ( P<0 . 020 ) during co-challenge infections when co-inoculated 1∶1 with wild-type ( Fig . 3B ) . With the exception of talA , both E . coli and P . mirabilis share the same complement of the Entner-Duodoroff and pentose phosphate genes; therefore it is unlikely that redundancy of transaldolase in E . coli can account for the disparate requirements for these pathways between the two pathogens . In support of this , the edd , talA , talB , talAtalB , and gnd strains in E . coli and the edd , talB , and gnd strains in P . mirabilis all demonstrate similar growth in vitro and also to both wild-type parental strains during culture in LB medium and defined medium containing glucose as the sole carbon source ( S1 Fig . ) . The aerobic tricarboxylic acid ( TCA ) cycle has been proposed to be required for E . coli fitness during growth on gluconeogenic substrates present in the urinary tract [8] . Specifically , E . coli sdhB mutant bacteria have been shown to have fitness defects during UTI [8] , [42] , suggesting that the reductive TCA cycle may not be operating during host colonization . To better define the role for the TCA cycle during extraintestinal infection , mutants of E . coli and P . mirabilis lacking succinate dehydrogenase; sdhB , fumarate dehydratase ( fumarase ) ; fumC , and fumarate reductase; frdA were constructed and tested in competitive infections with wild-type E . coli CFT073 or P . mirabilis HI4320 , respectively . While both E . coli and P . mirabilis required TCA cycle reactions for fitness in vivo , sdhB was required for fitness only during cystitis ( bladder CFU ) in E . coli ( Fig . 4A ) and only during pyelonephritis ( kidney CFU ) in P . mirabilis ( Fig . 4B ) ( P>0 . 050 ) . It is generally believed that the production of reduced FADH2 during the conversion of succinate to fumarate by succinate dehydrogenase is avoided during fermentation by modification of the TCA cycle to an incomplete reductive pathway where fumarate conversion to succinate by fumarate reductase replaces succinate dehydrogenase activity . The loss of FrdA resulted in a fitness defect for P . mirabilis during infection of both the bladder and kidneys ( P>0 . 005 ) ( Fig . 4B ) . In contrast , E . coli frdA mutant colonization levels were indistinguishable from wild-type ( median CI = 1 . 0 ) in the kidneys and significantly outcompeted the parental CFT073 strain>50-fold during acute cystitis ( P = 0 . 024 ) ( Fig . 4A ) . Both E . coli and P . mirabilis required fumC , which functions to convert fumarate to malate; loss of FumC , however , resulted in a severe fitness defect for P . mirabilis during bladder and kidney infection ( CI<10−3 , P<0 . 005 ) , while the E . coli fumC mutant colonized the kidneys to similar levels as the parental CFT073 strain ( CI = 0 . 94 ) and had a minor fitness defect in the bladder ( CI = 0 . 1 , P = 0 . 031 ) ( Fig . 4A ) . During bacterial growth on gluconeogenic substrates , peptides and certain amino acids that are present in the urinary tract are broken-down into pyruvate , which can be oxidized in the TCA cycle or reduced to fermentative end-products . The resulting oxaloacetate can fuel gluconeogenesis as the substrate for pyruvate carboxykinase ( pckA ) that generates phophoenolpyruvate and bypasses the irreversible glycolytic reaction catalyzed by pyruvate kinase ( pykA ) . Mutation of pckA , which disrupts gluconeogenesis , resulted in a significant fitness defect for E . coli in both bladder and kidneys ( P<0 . 005 ) ( Fig . 4A ) . Loss of pckA in P . mirabilis did not significantly affect colonization during urinary tract infection ( Fig . 4B ) . In support of differential utilization of amino acids present in the urinary tract , arginine and serine auxotrophs of E . coli demonstrate no fitness defect during UTI [8] , while in P . mirabilis , serine auxotrophy created a 100-fold decrease in bladder colonization ( CI = 10−2 , P<0 . 005 ) ( S2 Fig . ) . The variable requirement for gluconeogenesis between both pathogens was not due to differences during in vitro culture; pckA mutant strains in both CFT073 and HI4320 backgrounds were indistinguishable from parental strains in LB medium and defined medium containing glucose ( Fig . 4C and D ) . In E . coli , the sdhB mutant demonstrated a growth defect in LB medium ( Fig . 4C ) but not when cultured in defined glucose medium ( Fig . 4E ) , while the P . mirabilis frdA mutant demonstrated a growth defect in defined glucose medium ( Fig . 4F ) but not when cultured in LB medium ( Fig . 4D ) . Both E . coli and P . mirabilis with mutations in fumC demonstrated an in vitro growth defect in LB medium , but only the P . mirabilis fumC mutant was unable to replicate in defined glucose medium ( Fig . 4F ) . Although both of these pathogenic isolates require components of the TCA cycle for fitness during infection , these in vivo and in vitro data suggest a key difference in respiration during growth on glycolytic substrates exists between E . coli and P . mirabilis despite both being enteric bacteria . The striking difference in the central pathway requirements during UTI between E . coli and P . mirabilis are puzzling because the central pathways are conserved and both pathogens are being assessed for fitness in the identical ascending UTI model . This suggests that an activity associated with the growth of the bacteria may cause alterations in the nutrient availability within the urinary tract . To test this , we performed mixed infections where the same mutation was tested against the opposite wild-type isolate . It was possible to distinguish E . coli from P . mirabilis by performing viable counts on agar with and without tetracycline due to P . mirabilis innate TetR phenotype . We chose to test the gnd mutants in both E . coli and P . mirabilis because that mutation created a severe fitness defect in P . mirabilis at 7 days in the bladder and kidneys , while no effect on fitness was observed for E . coli at 48 h in either tissue ( Fig . 2A and B ) . In addition , the P . mirabilis gnd mutant demonstrated a fitness defect at 48 h in both the bladder and kidneys ( Fig . 5A ) . Surprisingly , we found that when mixed 1∶1 with wild-type E . coli CFT073 , the P . mirabilis gnd mutant was not out-competed at 48 h in either bladder or kidneys ( Fig . 5B ) . Further , the E . coli gnd mutant was now significantly out-competed by the wild-type P . mirabilis HI4320 by>100-fold in both the bladder and kidneys ( Fig . 5C ) . When the gnd mutants of each strain were mixed 1∶1 , there was no observable difference in competitive indices ( Fig . 5D ) . The same apparent reversal of in vivo fitness was also observed at 7 days post-inoculation ( Fig . 5E ) , while no in vitro growth advantage was observed in any combination of gnd mutant bacteria and wild-type E . coli or P . mirabilis ( Fig . 5F ) . Similarly , E . coli and P . mirabilis wild-type demonstrate equivalent growth during co-culture in LB medium and in defined medium with glucose as the sole carbon source ( S3 Fig . ) . The observed differences in central pathways requirements between E . coli and P . mirabilis during colonization of the urinary tract led us to speculate that both species could co-exist within the same urinary tract without directly competing for nutrients . By performing mixed inoculations of E . coli and P . mirabilis , it was also possible to test whether the observed fitness reversal of the gnd mutant bacteria could be explained by intrinsic differences in the level of colonization during UTI . Indeed , as expected , the CFU/g of bladder and kidneys at 7 d are 2–3 logs higher for P . mirabilis than E . coli during single strain infection ( Fig . 6A and B ) . When the wild-type E . coli CFT073 and P . mirabilis HI4320 were co-inoculated , however , the level of E . coli colonization in the bladder and kidneys increased by over 3 logs at both 48 h and 7 days post-inoculation ( Fig . 6C and D ) with a concomitant 10-fold increase in P . mirabilis colonization in both the bladder and kidneys ( Fig . 6D ) . To further test the compatibility of these uropathogens we also performed sequential infections . Mice were inoculated with a single strain and colonization was established for 48 h prior to infecting with the second strain . In these sequential infections we observed that pre-colonization of the urinary tract did not exclude colonization by the second strain ( S4 Fig . ) . Further , when P . mirabilis is used to infect and colonize mice , followed by infection with E . coli we observed enhanced colonization as seen when both bacteria are simultaneously co-inoculated into mice ( S4 Fig . ) . These data demonstrate that mixed infection provides an obvious benefit for E . coli during UTI , but also that the presence of E . coli provides a mutual benefit by allowing P . mirabilis to colonize to a greater density than it would by itself .
The primary objective for microorganisms is to grow or replicate . For pathogenic microorganisms this need to replicate is paramount for their ability to successfully colonize , establish infection and cause disease . Many bacteria have evolved specific pathways that provide a growth advantage in a specific nutritional niche . These specific pathways often involve transport systems that aid the bacterium to acquire a certain nutrient , such as the ability of uropathogenic E . coli to import and utilize D-serine [31] , [43] or the ability to sense α-ketoglutarate levels [44] . In contrast to these specific types of adaptations , most bacteria share highly conserved central pathways colloquially referred to as central metabolism . Our previous work began the comprehensive characterization of central pathways required by E . coli during UTI and found that gluconeogenesis and the TCA cycle were required to catabolize the dilute mixture of amino acids and peptides found in the urinary tract , while glycolysis , the pentose phosphate pathway , and the Entner Duodoroff pathway were dispensible [8] . In the present study , we extended this comprehensive study of E . coli central pathway requirements during UTI and in addition , performed parallel experiments using P . mirabilis , another well-studied uropathogen . We reasoned that both species of bacteria would have similar central metabolism requirements because they both occupy the same host niche , the urinary tract . Unexpectedly , we found that E . coli and P . mirabilis have strikingly divergent central pathway requirements despite infecting and growing in an identical host environment with presumably access to the same nutrients . Our finding that E . coli and P . mirabilis have different central pathway requirements suggests that either a specific activity of the bacteria alters the host niche or that there is an intrinsic difference in the metabolic capabilities between the bacteria . Since we are studying highly conserved central pathways that are present in both species it is reasonable to conclude that a specific activity is present in one that is lacking in the other species . When considering the host urinary tract , the available nutrients are amino acids , peptides , and urea [40] . One obvious difference between Proteus and E . coli is that the former produces urease enzyme [45] , [46] , which hydrolyzes urea into ammonia and carbon dioxide ( Fig . 7 ) . The presence of urease activity would create a nitrogen rich environment by vastly increasing nitrogen availability via the concomitant increased ammonia concentration . In turn , the C/N ratio would be dramatically reduced for a urease producing bacteria like P . mirabilis relative to E . coli that does not have genes to encode urease . The altered C/N ratio would have profound effects on central pathway utilization because carbon metabolism and nitrogen assimilation is highly integrated [47] , [48] , [49] . Indeed , the apparent divergence in the ability to sense available nitrogen in urea results in E . coli activation of the glutamine synthetase and glutamate oxo-glutarate aminotransferase system ( GS/GOGAT ) to assimilate nitrogen [34] , [35] , while P . mirabilis assimilates nitrogen , via glutamate dehydrogenase ( Gdh ) [36] due to the apparent excess nitrogen available from ammonia produced by urea hydrolysis . We reason that this key difference is partly responsible for P . mirabilis requiring glycolysis , pentose phosphate pathway , and the ED pathway; while , on the other hand , the exact same mutations in E . coli have no affect on fitness during UTI . Alternatively , our findings raise the possibility that E . coli and P . mirabilis could reside in different cellular compartments . For example , it has been shown that E . coli can reside intracellularly during acute infection [50] , [51] , and while P . mirabilis can invade cultured cells [52] , [53]; it is unclear if P . mirabilis spends a significant portion of the acute infection within host cells . In addition to finding a remarkable difference in central metabolism requirements between two pathogens that infect the urinary tract , these findings have validated that glycolysis is dispensable for E . coli during UTI . In our previous work we assessed glycolysis by studying a mutation in the gene that encodes triose phosphate isomerase [8]; however , that enzyme is reversible . In the present study we created and tested phophofructokinase- and pyruvate kinase-deficient mutants in E . coli and P . mirabilis . These enzymes perform irreversible steps in glycolysis , thus by testing these mutations during UTI it is now clear that glycolysis is dispensible for E . coli and is required for P . mirabilis fitness in vivo . That glycolysis is required for P . mirabilis but not E . coli raises the possibility that sugars become available within the urinary tract during Proteus infection . That sugars become available when P . mirabilis colonizes the urinary tract would also allow for enhanced colonization of E . coli because sugars available during co-infection might increase growth of E . coli over numbers it would normally reach by solely consuming amino acids . This could also explain why gnd mutant E . coli displayed a fitness defect only when co-colonized with P . mirabilis . It has also been shown that the oxidative TCA cycle is important for E . coli fitness during UTI [8] , [42] , but whether or not the reductive TCA cycle is important for fitness during UTI remains unanswered . We found that mutants lacking fumarate reductase , which is an enzyme that utilizes fumarate as an electron acceptor during anaerobic respiration , do not have a fitness defect in E . coli as predicted by our earlier study . In contrast , fumarate reductase mutants in P . mirabilis are out-competed by wild-type . These data show that the branched , reductive TCA pathway is not important for E . coli to grow within the urinary tract , supporting the notion that the urinary tract is moderately oxygenated [35] . However , the result that fumarate reductase is important for P . mirabilis fitness during UTI suggests an anaerobic environment does exist in the host urinary tract . In addition , because sensing oxygen depletion has been shown to be important for E . coli fitness during UTI [44] , it remains possible that fumarate reductase is dispensible for E . coli due to the presence of alternative energy pathways like Ni-Fe hydrogenases and nitrate reductase , these and other redox enzymes are part of the large repertoire of electron transfer components available for the highly modular respiratory chains that can be assembled in E . coli [54] . Future studies will be useful to determine the exact energy pathways used by E . coli during UTI . The urogenital track of humans is considered a site of polymicrobial colonization [55] and it has been shown that UTI can be polymicrobial [56] , including co-infections with E . coli and P . mirabilis [57] , [58] . Polymicrobial UTI has also been suggested to enhance E . coli virulence [58] . Our findings that two preeminent UTI pathogens have divergent central metabolism requirements have important implications for polymicrobial infections . Specifically , these data indicate that E . coli and P . mirabilis may not directly compete for nutrients during colonization of the urinary tract and led us to hypothesize that co-inoculation of E . coli and P . mirabilis may lead to enhanced colonization levels during experimental UTI . Indeed , we found that colonization was enhanced for both E . coli and P . mirabilis . The benefit of polymicrobial infection was more apparent for E . coli over time . In addition to supporting that E . coli and P . mirabilis may not directly compete for resources during UTI , these findings could alternatively suggest that some activity of one or both pathogenic bacteria may alter the host niche in such a way that facilitates bacterial replication . In support of this , we observed that P . mirabilis causes more tissue damage and larger induction of the pro-inflammatory cytokines IL-1α , IL-β , IL-6 , and G-CSF than does E . coli in independent infection , while a mixed infection appears more similar to E . coli alone ( S5 Fig . ) . With a systematic view of bacterial metabolism during UTI in hand , further studies of the host innate response to these infections will help form a comprehensive view of how the host response shapes bacterial carbon utilization during UTI . To test whether or not the bacteria differences in the host urinary tract niche , we decided to co-inoculate E . coli and P . mirabilis mutants with the heterologous wild-type strain and assess fitness during UTI . Our findings showed that polymicrobial infection of the urinary tract changed the fitness results when compared to mono-species infection . The E . coli gnd mutant , which was not outcompeted by its parent strain , demonstrated a colonization disadvantage when co-inoculated with P . mirabilis wild-type . Conversely , the P . mirabilis gnd mutant , which was out-competed by its parent strain , did not demonstrate a disadvantage in colonization when co-inoculated with wild-type E . coli . Although future studies with additional mutants will be useful to further elucidate how co-inoculation might change central pathway requirements , these data do suggest that the host niche environment is altered by bacterial activity during UTI . Importantly , our study leads to a more complete picture of the metabolism of two key bacterial pathogens that cause UTI and shows that polymicrobial infection of the urinary tract may alter the metabolic pathways required for optimal growth within the host . These findings provide a better understanding of bacterial metabolism during clinically relevant infections and represent an important foundation to begin to dissect how metabolism and virulence intersect during UTI and how polymicrobial interactions may affect pathogenesis of extraintestinal E . coli infections .
P . mirabilis HI4320 was isolated from urine of a patient presenting with bacteriuria during long-term catheterization [56] , [59] . E . coli CFT073 was isolated from the blood and urine of a patient with acute pyelonephritis [37] . E . coli and P . mirabilis were routinely cultured in lysogeny broth ( LB ) medium . For growth experiments , wild-type and mutant strains of E . coli and P . mirabilis were cultured in MOPS defined medium [60] and minimal salts medium [61] , respectively , both containing either 0 . 2% ( w/v ) glucose or 0 . 2% ( w/v ) glycerol as the sole carbon source . Defined medium cultures were inoculated 1∶50 and LB cultures were inoculated 1∶100 from overnight LB bacterial cultures and incubated with aeration at 37°C . Growth curves were performed in triplicate; OD600 was recorded every hour . P . mirabilis HI4320 mutants ( S1 Table ) were generated using the TargeTron Gene Knockout System ( Sigma ) . Oligonucleotides for mutant construction were created using the TargeTron Design Site ( Sigma ) . PCR confirmation of mutants was performed using oligonucleotides flanking the intron insertion site designed with the PrimerQuest program on the Integrated DNA Technologies website . E . coli CFT073 deletion mutants ( S1 Table ) were constructed using the lambda red recombinase system [62] . Primers homologous to sequences within the 5′ and 3′ ends of the target genes were designed and used to replace target genes with a nonpolar kanamycin- or chloramphenicol-resistance cassette derived from the template plasmid pKD4 or pKD3 , respectively [62] . Confirmation of E . coli mutants was carried out by PCR using primers flanking the target gene sequence and comparing product size to wild-type PCR product size . When size differences were negligible PCR products were digested with a restriction enzyme ( New England Biolabs ) . For in vitro complementation , the tpiA gene was amplified from P . mirabilis genomic DNA using Easy-A high fidelity polymerase ( Stratagene ) and independently cloned into pGEN-MCS [63] , [64] using appropriate restriction enzymes . The sequences of pGEN-tpiA were verified by DNA sequence analysis prior to electroporation into the P . mirabilis tpiA mutant strain and E . coli tpiA mutant strain . The CBA mouse model of ascending UTI [39] , [65] was used to assess the fitness contribution of each metabolic mutant during co-challenge competition . To determine persistence of wild-type strains , independent infections of a single strain were performed . Female CBA/J mice ( 6–8 week old; 20 to 22 g; Jackson Laboratories ) were anesthetized with ketamine/xylazine and transurethrally inoculated with a 50 µl bacterial suspension ( total inoculum = 5×107 or 2×108 CFU ) per mouse using a sterile polyethylene catheter ( I . D . 0 . 28 mm × O . D . 0 . 61 mm ) connected to an infusion pump ( Harvard Apparatus ) . For in vivo co-challenges , a suspension containing 5×107 CFU of a 1∶1 ratio of P . mirabilis HI4320 and P . mirabilis kanamycin-resistant mutant in LB medium or a suspension containing 2×108 CFU of a 1∶1 ratio of E . coli CFT073 and E . coli antibiotic-resistant mutant in PBS . For independent infections , the respective P . mirabilis or E . coli suspensions contained only the wild-type strain . Input CFU/ml was determined by plating serial dilutions ( Spiral Biotech ) of each inoculum onto low salt ( 0 . 5 g NaCl/L ) LB agar , to prevent P . mirabilis swarming , with and without antibiotic . For experiments with P . mirabilis , low salt LB agar ( 0 . 5 g NaCl/L ) was used to prevent swarming . Infected mice were euthanized 48 h or 7 d post infection , bladder and kidneys were aseptically removed , weighed , and homogenized ( OMNI International ) in 3 ml PBS , and appropriate dilutions were spiral plated on LB agar with and without antibiotic to determine the output CFU/g of tissue . Viable counts were enumerated using QCount software ( Spiral Biotech ) and CFU from antibiotic-containing medium ( mutant CFU ) were subtracted from the total CFU from plates lacking antibiotic to determine the number of wild-type bacteria . For co-challenge experiments , competitive indices ( CI ) were calculated by dividing the ratio of the CFU of mutant to wild-type recovered from each mouse following infection by the ratio of the CFU of mutant to the CFU of wild-type present in the input . CI data were log-transformed and analyzed by the Wilcoxon signed-rank test to determine statistically significant differences in colonization ( P-value <0 . 05 ) . A CI>1 indicates that the mutant out-competes the wild-type strain and a CI<1 indicates that the mutant is out-competed by the wild-type strain . For independent infections the Mann-Whitney test was used to determine statistically significant differences in colonization ( P-value <0 . 05 ) . The relative fitness in vivo for bacteria during polymicrobial infection was determined by co-inoculating UPEC CFT073 and P . mirabilis HI4320 strains and deletion mutants into the same female CBA mice as described previously with the following modification . For polymicrobial co-challenge infections , a 1∶1 ( v/v ) mixture was prepared containing 2 . 5×107 CFU of P . mirabilis HI4320 in LB medium and 108 CFU E . coli CFT073 in PBS . Competitive indices were calculated as described above . For polymicrobial infections containing only wild-type strains , the CFU/g tissue were determined following plating of serial dilutions on low salt ( 0 . 5 g NaCl/L ) LB agar with and without tetracycline; P . mirabilis is intrinsically tetracycline-resistant . CFU from tetracycline agar plates , which represent P . mirabilis , were subtracted from total CFU recovered on LB agar without antibiotics to determine CFU/g for E . coli . For quantification of bacteria recovered from a polymicrobial infection with a wild-type strain and a heterologous kanamycin-resistant mutant strain , CFU on LB agar containing kanamycin ( mutant ) were subtracted from total CFU recovered on LB without antibiotics , to determine wild-type CFU/g tissue . Tetracycline was used to enumerate bacterial colonization levels following polymicrobial infection with heterologous kanamycin-resistant strains . All animal experiments were performed in accordance to the protocol ( 08999-3 ) approved by the University Committee on Use and Care of Animals at the University of Michigan . This protocol is in complete compliance with the guidelines for humane use and care of laboratory animals mandated by the National Institutes of Health . | The human urinary tract is a leading source for polymicrobial infections and for the development of bacteremia and sepsis . Treating these potentially dangerous infections have recently become more challenging due to the appearance of uropathogenic strains that are resistant to the many of the most commonly prescribed antibiotics . The majority of urinary tract infections ( UTI ) are caused by Escherichia coli , while another bacterium , Proteus mirabilis , is more likely to cause catheter-associated UTI . Here , we report that uropathogenic E . coli and P . mirabilis have divergent nutritional requirements despite growing in the same host environment . This result indicates that E . coli and P . mirabilis do not directly compete for nutrients during UTI . Indeed , we found that persistence of both pathogens is enhanced when they co-colonize the host . This work represents an important step toward understanding the basic nutritional requirements for two major pathogens that cause UTI and shows how mixed infections can change these requirements . Understanding how bacteria grow during infections is fundamental to ultimately uncover new ways to combat increasingly drug-resistant bacterial infections . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"microbial",
"metabolism",
"medicine",
"and",
"health",
"sciences",
"microbial",
"physiology",
"pathology",
"and",
"laboratory",
"medicine",
"host-pathogen",
"interactions",
"medical",
"microbiology",
"microbial",
"pathogens",
"biology",
"and",
"life",
"sciences",
"microbiology",
"pathogenesis",
"bacterial",
"pathogens"
] | 2015 | Preferential Use of Central Metabolism In Vivo Reveals a Nutritional Basis for Polymicrobial Infection |
Schistosomiasis , caused by infection with the blood fluke Schistosoma , is responsible for greater than 200 , 000 human deaths per annum . Objective high-throughput screens for detecting novel anti-schistosomal targets will drive ‘genome to drug’ lead translational science at an unprecedented rate . Current methods for detecting schistosome viability rely on qualitative microscopic criteria , which require an understanding of parasite morphology , and most importantly , must be subjectively interpreted . These limitations , in the current state of the art , have significantly impeded progress into whole schistosome screening for next generation chemotherapies . We present here a microtiter plate-based method for reproducibly detecting schistosomula viability that takes advantage of the differential uptake of fluorophores ( propidium iodide and fluorescein diacetate ) by living organisms . We validate this high-throughput system in detecting schistosomula viability using auranofin ( a known inhibitor of thioredoxin glutathione reductase ) , praziquantel and a range of small compounds with previously-described ( gambogic acid , sodium salinomycin , ethinyl estradiol , fluoxetidine hydrochloride , miconazole nitrate , chlorpromazine hydrochloride , amphotericin b , niclosamide ) or suggested ( bepridil , ciclopirox , rescinnamine , flucytosine , vinblastine and carbidopa ) anti-schistosomal activities . This developed method is sensitive ( 200 schistosomula/well can be assayed ) , relevant to industrial ( 384-well microtiter plate compatibility ) and academic ( 96-well microtiter plate compatibility ) settings , translatable to functional genomics screens and drug assays , does not require a priori knowledge of schistosome biology and is quantitative . The wide-scale application of this fluorescence-based bioassay will greatly accelerate the objective identification of novel therapeutic lead targets/compounds to combat schistosomiasis . Adapting this bioassay for use with other parasitic worm species further offers an opportunity for great strides to be made against additional neglected tropical diseases of biomedical and veterinary importance .
Infection with the parasitic trematode Schistosoma mansoni causes a wide range of quantifiable clinical pathologies [1] , which collectively lead to the death of approximately 200 , 000 individuals/annum [2] . Recent ‘first pass’ description of this parasite's genome [3] , as well as multiple reports describing the utilization of numerous functional genomics tools ( e . g . [4] , [5] ) , have now provided the technological framework for a renaissance in drug target and vaccine discovery research [6] . A major bottleneck in converting schistosome phenotypic discovery into applied therapeutic products , however , is the lack of appropriate methods for quantifying , in a high-throughput manner , individual gene function or small compound effect on parasite survival . Therefore , development of reproducible , non-subjective methods for high-throughput screening of parasite viability would present the schistosome community with a tangible opportunity to translate genomic and functional genomics information into therapeutic strategies to combat schistosomiasis . Current methods utilized to assess schistosome and other trematode viability have recently been reviewed [7] . All involve microscopic techniques where the experimenter manipulates the parasite in vitro and assesses the effect of such manipulation by bright-field examination of morphology . This technique has been employed in immunological studies [8] , RNA interference ( RNAi ) assays [9] , drug screening protocols [9] , [10] and general manipulations of parasite development [11] . Criteria used to assess schistosome viability in these investigations include intracellular granularity , parasite movement , parasite shape alterations and uptake of various vital dyes ( e . g . methylene blue or toluidine blue ) . The subjective nature of these various and time-consuming measurement indices indicate that inter-laboratory estimates of schistosome viability in response to in vitro manipulation will be quite variable , resulting in a lack of uniform reporting within the community . We report here on an improved methodology to objectively detect parasite survival during in vitro culture . The fundamental principle of this assay is derived from the differential membrane permeability of two well-known dyes , fluorescein diacetate ( FDA , an esterase substrate ) and propidium iodide ( PI , a DNA intercalating agent ) . Based on previously described uses of these two dyes [12] , it was anticipated that FDA would cross the membranes of living cells ( within living schistosomes ) and be converted into highly-fluorescent and charged fluorescein ( which cannot readily leave live cells ) by parasite esterase activity . In contrast to FDA , PI would not be able to cross the membranes of living parasites and could only stain nucleic acids if there was a breach in membrane permeability ( due to parasite death ) . In dead parasites , we hypothesized that PI , and not FDA , would preferentially stain the assayed sample and , thus , changes in both dye's fluorescent intensities would be indicative of schistosome viability , which could subsequently be quantified by a plate reader , fluorescent microscope or any device equipped to measure fluorescence . This single property ( simultaneous detection of both PI and FDA measures ) has allowed us to develop a fluorescence-based , microtiter-plate bioassay to improve detection of schistosome viability , which is high-throughput ( 96- and 384-well capacity ) , quantitative and provides objective readouts ( fluorescence intensity units ) of parasite survival during in vitro culture . Using this flexible bioassay , we demonstrate its versatility in detecting schistosome survival in response to thioredoxin glutathione reductase ( TGR ) inhibition [9] . Furthermore , we adopt this assay to provide quantitative estimates of schistosome viability in the presence of recently identified small compounds with previously described [10] and unknown [3] chemotherapeutic activities . Implementation of this novel screening platform by academia and industrial stakeholders will enable inter-laboratory comparisons of in vitro parasite manipulations to be routinely and quickly performed , vastly accelerating the search for novel anti-schistosomal lead targets .
Schistosoma mansoni ( Puerto Rican strain ) infected Biomphalaria glabrata snails were provided by Fred Lewis ( Biomedical Research Institute , Rockville , MD , USA ) . Cercariae were shed from infected snails by exposure to light ( 60 min at room temperature , RT ) and subsequently converted to schistosomula by mechanical transformation [13] . Schistosomula were purified away from cercarial tails by centrifugation through a 60% percoll gradient [14] . Microscope examination was used to assess the quantity and quality of purified schistosomula . Schistosomula were cultured at 37°C in T25 tissue culture flasks containing 9 ml DMEM ( Dulbecco's Modified Eagle Medium , Sigma-Aldrich ) , lacking phenol red but containing 4500 mg/l glucose , supplemented with 10% foetal calf serum , 2 mM L-glutamine , 200 U/ml penicillin , 200 µg/ml streptomycin ( all Sigma-Aldrich ) in an atmosphere of 5% CO2 for 24 hr before any further experimental manipulations proceeded . Negligible parasite death occurred in this media during the 24 hr culturing period . Following this , schistosomula were aliquoted into black-sided , flat-bottom ( optically clear ) , 96-well microtiter plates ( Fisher Scientific ) in 200 µl media or black-sided , flat-bottom ( optically clear ) , 384-well microtiter plates ( Matrix ) in 40 µl media . Heat killed schistosomula were also prepared by incubating the 24 hr cultivated parasites at 65°C for 10 min . These dead schistosomula were allowed to cool to 37°C before being used in subsequent experiments . Live and heat killed schistosomula stained with optimal concentrations ( empirically derived from [12] ) of propidium iodide ( PI , 2 . 0 µg/ml; Sigma-Aldrich ) , fluorescein diacetate ( FDA , 0 . 5 µg/ml; Sigma-Aldrich ) or both fluorophores were visualized at ×100 magnification using a Leica Axioplan microscope equipped with FITC ( 494 excitation ) and Rhodamine ( 536 excitation ) filters and a mercury vapor light source . A Hamamatsu CA74295 camera with Wasabi Version 1 . 4 software was used to capture photographic images of stained schistosomula . Schistosomula , co-incubated with test compounds , were fluorescently visualized as above or unstained at ×100 magnification using an Olympus CK2 inverted microscope equipped with a stage extension plate and specimen holder for handling microtiter plates . A Kodak EasyShare DX7440 digital camera was used to capture images of unstained schistosomula . All compounds were purchased from Sigma-Aldrich and included: gambogic acid , sodium salinomycin , ethinyl estradiol , fluoxetidine hydrochloride , bepridil , ciclopirox , miconazole nitrate , chlorpromazine hydrochloride , amphotericin b , niclosamide , rescinnamine , flucytosine , vinblastine , carbidopa , praziquantel and auranofin . Stock solutions of all compounds were made up at 1 mM in appropriate solvents ( Dataset S1 ) and stored at −80°C . All compounds were added to black-sided , flat-bottom ( optically clear ) , 96-well microtiter plates containing schistosomula ( 1000 parasites/well in triplicate ) at 10 µM concentrations . Schistosomula were cultured ( as already indicated; 37°C , 5% CO2 ) in the presence of each compound for 24 hr before viability levels were assessed . After the 24 hr culturing period in the presence of test compounds , and prior to addition of fluorescent dyes , all schistosomula were washed three times to remove test compound and culture media supplements . Each wash consisted of centrifuging microtiter plates containing schistosomula at 100×g for 5 min , removal of half the old culture media and replacement with an equal quantity of fresh DMEM ( lacking phenol red ) . After washing the parasites , PI and FDA were simultaneously added to each well of the microtiter plate to obtain a final concentration of 2 . 0 µg/ml and 0 . 5 µg/ml respectively . The 96-well microtiter plates , now containing fluorescently labeled parasites , were subsequently loaded into a BMG Labtech Polarstar Omega plate reader containing appropriate filters for the simultaneous detection of PI ( 544nm excitation/620nm emission ) and FDA ( 485nm excitation/520nm emission ) . All fluorescent values were obtained with the plate reader incubator set at 37°C to ensure efficient esterase conversion of FDA to fluorescein within live schistosomula . The plate reader automatically sets the photo multiplier tube ( PMT ) gain for each fluorescent dye and this may slightly vary between experiments . Inclusion of appropriate control samples ( live and dead schistosomula ) compensates for any inter-plate variations in gain settings . All data were exported into Microsoft Excel for organization and into Minitab ( Version 14 ) for statistical analyses . A One-Way Analysis of Variance ( ANOVA ) followed by post hoc testing with Fisher's least significant difference ( LSD ) was used to detect statistical differences between treatments . Viability percentages were either: ( 1 ) converted into probits for auranofin dose response curve generation and calculation of LD50 values or ( 2 ) Arcsin transformed in order to stabilize variances prior to the use of appropriate statistical analyses . Numbers of live and dead schistosomula in each well of a microtiter plate were calculated using the following equations:‘Samples’ represent fluorescence intensity units collected from parasites incubated with test compounds . ‘Negative control’ represents fluorescence intensity units collected from parasites killed with 10µM auranofin or heat shock ( 10 min incubation at 65°C ) , while the ‘Positive control’ represents fluorescence intensity units collected from untreated parasites . ‘Media control’ represents fluorescence intensity units originating from wells containing only media ( no parasites ) . To determine schistosomula viability in response to each test compound , a second calculation was employed . This normalization compensated for inter-well variability in schistosomula numbers across the microtiter plate , and to facilitate accurate viability comparisons between test compounds from different microtiter plates:
To first determine whether propidium iodide ( PI ) and fluorescein diacetate ( FDA ) could be used individually to detect dead or live in vitro transformed S . mansoni schistosomula , an initial fluorophore uptake experiment was performed with parasite viability assessed by fluorescent microscopy . Here , in vitro transformed schistosomula , cultured for 24 hr , were either heat killed at 65°C for 10 min or left undisturbed at physiological conditions ( 37°C ) . Heat-killed parasites were then subsequently single-stained with PI ( 2 . 0 µg/ml ) whereas live parasites were incubated with FDA ( 0 . 5 µg/ml ) . All dead parasites fluoresced ( Fig . 1A , 16 out of 16 ) when visualized for PI uptake at 536 nm , with corresponding polarized bright field microscopy imaging of the same schistosomula samples providing morphological confirmation of death ( i . e . parasites that were uniform in shape and size and displayed no movement ( Fig . 1B ) ) . Furthermore , all live parasites fluoresced ( Fig . 1C , 10 out of 10 ) when visualized for FDA uptake at 494 nm , with corresponding polarized bright field microscopy imaging of the same schistosomula samples providing visual confirmation of a physiological normal phenotype ( i . e . parasites displaying a variety of shapes and sizes as the result of movement during in vitro culturing , Fig . 1D ) . Further quantification of fluorophore uptake was performed on four different regions of slides containing PI stained schistosomula and from five different regions of slides containing FDA stained schistosomula . Here 100% of the parasites were stained with either of the two examined fluorophores ( 39 out of 39 dead schistosomula were PI positive and 48 out of 48 live schistosomula were FDA positive ) , confirming the utility of the chosen fluorophores for evenly staining schistosomula ( data not shown ) . By demonstrating that PI could effectively be used to stain dead schistosomula and FDA was capable of identifying live parasites , a further experiment was conducted to determine whether PI and FDA could simultaneously be used to detect individual live ( FDA positive ) and dead ( PI positive ) in vitro transformed schistosomula from a mixed population of differentially viable parasites . Here , in vitro transformed schistosomula , cultured for 24 hr , were either heat killed at 65°C for 10 min or left undisturbed at physiological conditions ( 37°C ) . Equal numbers of live and dead schistosomula were mixed , simultaneously stained with PI ( 2 . 0 µg/ml ) and FDA ( 0 . 5 µg/ml ) and examined for the presence or absence of differential fluorophore emission . Whereas heat-killed schistosomula exhibited strong PI staining ( minimal FDA staining ) , physiologically normal schistosomula conversely displayed strong FDA staining ( no observable PI staining ) ( compare Fig . 1E and Fig . 1F ) . This differential staining of individual live and dead parasites was further supported by observations of undetectable co-localization of PI and FDA signals ( within the same cell of an individual schistosomulum , detectable as a yellow signal ) , schistosomula phenotype and parasite motility as detected by both fluorescent and polarized bright field microscopy ( Fig . 1G and 1H ) . A critical first parameter in translating these observations of differential PI and FDA staining of schistosomula to a method that takes advantage of the high-throughput potential of a microtiter plate assay ( in both 96- and 384- well formats ) was to determine the optimal timeframe that each fluorescent dye should be incubated with parasite samples to insure maximal reproducible detection of viability ( Fig . 2 ) . Here , in vitro transformed schistosomula , cultured for 24 hr , were either killed by heat shock ( 65°C for 10 min ) or left unaltered at physiological culturing conditions ( 37°C ) . Three samples were subsequently derived from these schistosomula cultures , which included physiologically normal , untreated parasites ( Fig . 2 , green line ) , heat-killed parasites ( Fig . 2 , red line ) and an equal mixture of physiologically normal and heat-killed parasites ( Fig . 2 , brown line ) . Derived schistosomula samples ( in triplicate ) were co-stained with PI ( 2 . 0 µg/ml ) and FDA ( 0 . 5 µg/ml ) and subjected to fluorescent intensity measurements ( BMG Labtech Polarstar plate reader ) every minute for 120 min . Fluorescent intensity values derived from triplicate wells containing DMEM ( lacking phenol red ) , PI ( 2 . 0 µg/ml ) and FDA ( 0 . 5 µg/ml ) were also obtained over this timeframe ( Fig . 2 , dotted line ) . Throughout the assayed 120 min timeframe , the three tested schistosomula samples ( regardless of plate format ) demonstrated clear differences in PI fluorescence emission , with the dead parasites demonstrating the highest PI values , the live parasites demonstrating the lowest PI values and the mixed population of live and dead parasites demonstrating intermediate PI values between the extremes ( Figs . 2A and 2B ) . While statistical significance in measured PI fluorescent emission continued to increase ( i . e . p<0 . 05 at 1 min and p<0 . 001 for any time point after 4 min ) with time for any of the compared schistosomula samples , PI fluorescent emission differences between physiologically normal schistosomula ( live ) and DMEM ( lacking phenol red ) samples remained small throughout the duration of the 120-minute timeframe and never reached statistical significance ( p<0 . 001 ) . Therefore , based on these analyses , the optimal PI incubation time to distinguish between dead and live schistosomula is between 4 and 120 min , regardless of plate format . We chose 20 min to collect PI data as it provided an adequate time window to process multiple microtiter plates ( indicated in Fig . 2A and 2B ) and was well within the calculated window of accurately being able to determine statistical significance amongst live and dead schistosomula . Whereas PI staining of parasite samples yielded differential fluorescent results over the entire 120-minute timeframe ( Fig . 2A and 2B ) , FDA staining of live , dead and mixed schistosomula populations in either 96-well or 384-well microtiter plate formats generated fluorescent data that quickly reached a plateau ( 51 min for 96-well microtiter plates , Fig . 2C and 32 min for 384-well microtiter plates , Fig . 2D ) . Therefore , collection of FDA fluorescence was halted at 51 min . Nonetheless , FDA fluorescent intensity units measured across these time intervals produced data as expected with live parasites fluorescing brightest , dead parasites weakest and mixed live/dead parasite populations intermediate between the two ( Fig . 2C and 2D ) . Furthermore , differences in detected FDA fluorescence between any of the three schistosomula samples ( in both microtiter plate formats ) were found to be statistically significant between 3 min and 12 min . However , unlike the PI timecourse ( Fig . 2A and 2B ) , differences in FDA fluorescent emission originating from wells containing DMEM and all schistosomula samples were statistically significant beginning at 3 minutes ( p<0 . 05 ) . This finding makes DMEM unsuitable for use as a blank for detecting non-specific FDA fluorescence . Therefore , based on these analyses , the optimal FDA incubation time to distinguish between dead and live schistosomula in both microtiter plate formats is between 3 and 12 min . We chose 5 min ( indicated in Fig . 2C and 2D ) to collect FDA data in both microtiter plate formats as this fluorophore is prone to spontaneous hydrolysis [15] . To determine the sensitivity limits of this fluorescence-based assay in measuring schistosomula viability in a medium and a high throughput manner , two different experimental approaches were considered ( Figs . 3 and 4 ) . In the first approach ( Fig . 3A–D ) , schistosomula were serially diluted ( 5000 reducing to 36 parasites for a medium throughput 96-well plate format; 1000 reducing to 8 parasites for a high-throughput 384-well plate format , in triplicate ) to identify the absolute minimum number of parasites that could be reproducibly detected . Briefly , in vitro transformed schistosomula , cultured for 24 hr , were either heat killed ( 65°C for 10 min ) or left unaltered at physiological conditions ( 37°C ) . After serially diluting these differentially treated schistosomula samples , FDA or PI fluorophores were added and measurement of fluorescence intensity proceeded , within the parameters determined above ( 5 min for FDA and 20 min for PI ) . Correlation analysis revealed a strong linear relationship between PI fluorescence and dead schistosomula number . For the 96-well plate format ( Fig . 3A ) a high correlation ( r = 0 . 995 ) existed between the measured PI fluorescence and numbers of dead schistosomula sampled , while for the 384-well plates ( Fig . 3B ) a slightly lower correlation ( r = 0 . 986 ) existed between measured PI fluorescence and numbers of dead schistosomula . Correlation analysis of FDA fluorescence and live schistosomula also revealed a strong linear relationship ( r = 0 . 952 for the 96-well plate format , Fig . 3C and r = 0 . 977 for the 384-well format , Fig . 3D ) , although the linearity was affected by an FDA-associated fluorescence plateau effect seen when increased numbers of schistosomula were assayed . From these experiments , it was determined that optimal numbers of schistosomula for assays conducted in a 96-well microtiter plate are between 500 and 2500 parasites per well , while for assays conducted in a 384-well microtiter plate , optimal numbers of schistosomula are between 50 and 500 per well . The second approach used to interrogate the sensitivity of this fluorescence-based assay for detecting schistosomula viability was to replicate conditions where varying percentages of live and dead schistosomula would be found in the same sample ( i . e . in vitro drug assays , where the drug tested is less than 100% efficient in killing ) . By mixing different percentages of live ( physiologically normal ) and dead ( heat killed ) parasites ( Fig . 4 ) , we tested the ability of PI ( Figs . 4A and 4C ) and FDA ( Figs . 4B and 4D ) to distinguish viability in a population of 1000 schistosomula in a 96-well plate ( Fig . 4E ) and 200 schistosomula in a 384-well plate ( Fig . 4F ) . In both plate formats , intra-well PI fluorescence decreased and FDA fluorescence increased when greater percentages of live schistosomula were being examined . This made it possible to differentially identify schistosomula viability in intervals of 25% for both 96- and 384-well plate formats . However , the standard deviations reported for some values indicated that there was variability inherent in pipetting with discrete units of this size . The use of a dual staining method was , therefore , important for counteracting any pipetting errors and verifying that differences in fluorescence are genuinely caused by differences in mortality . To validate this fluorescent-based viability assay for practical application in medium to high-throughput drug screening , it was used to assess the efficacy of a known anti-schistosomula compound , auranofin ( Fig . 5 ) . An inhibitor of parasite thioredoxin glutathione reductase ( TGR ) , auranofin has been shown ( estimated by light microscopy examination ) to be 100% lethal to in vitro cultured , mechanically-transformed schistosomula at 10 µM concentrations , 24 hr after treatment [9] . In agreement with published reports , there is a clear and titratable anti-schistosomula effect mediated by auranofin with the dual fluorescent staining procedure allowing viability quantification of each drug concentration at 24 hours post-treatment ( Fig . 5A ) . Percent viability transformations into probit values also allowed an auranofin LD50 of 0 . 82±0 . 49 to be calculated ( Fig . 5B ) . Maximum drug effect was seen at 10 µM , where microscopic examination of schistosomula confirmed that death was 100% ( Fig . 5C , 5E and 5F ) , when compared to untreated parasites ( Fig . 5D ) . At an auranofin concentration within the calculated LD50 range ( 1 µM ) , schistosomula contained cells either labeled with PI or FDA throughout the lophotrochozoan body ( Fig . 5G and 5H ) . Statistically-significant , auranofin-mediated mortality , compared to either vehicle treated ( DMSO ) schistosomula or untreated parasites ( live ) , was also observed for drug concentrations of 5 µM , 2 . 5 µM and 1 . 25 µM . Further validation of this viability assay was next performed by assessing the effect of compounds with previously-identified [10] or suggested anti-schistosomal activities [3] ( Fig . 6 ) . These compounds included four ( gambogic acid , sodium salinomycin , niclosamide nitrate and amphotericin b ) that have been described to induce schistosomula mortality [10] , three ( ethinyl estradiol , fluoxetine hydrochloride and chlorpromazine hydrochloride ) that have been reported to induce schistosomula over-activity [10] , two ( miconazole nitrate and praziquantel ) that have been shown to produce a shape alteration ( ‘rounded’ phenotype ) [10] and six ( bepridil , ciclopirox , rescinnamine , flucytosine , vinblastine and carbidopa ) that have never been tested on schistosomula [3] ( full details in Dataset S1 ) . Of the tested compounds with previously-recorded effects on schistosome phenotypes [10] , our dual-fluorescent viability screen , reassuringly showed broad agreement ( Fig . 6A ) . Here , the compounds previously described as inducing an over-active ( ethinyl estradiol , fluoxetine hydrochloride and chlorpromazine hydrochloride ) or rounded ( miconazole nitrate and praziquantel ) phenotype , as expected , did not affect schistosomula viability . Fluorescent readings obtained from wells containing parasites treated with these compounds were no different from wells containing untreated parasites ( average viability 84 . 3% , data not shown ) . Microscopic examination of these treated parasites confirmed their viability ( e . g . Fig . 6E and Figure S1 ) . However , only two of four previously defined anti-schistosomula compounds ( gambogic acid and amphotericin b ) induced measurable death as determined by our dual-fluorescent viability assay ( confirmed by microscopic examination of schistosomula , e . g . Fig . 6B and Fig . S1 ) . Sodium salinomycin and niclosamide produced fluorescent viability measurements similar to those derived from untreated parasites . Detailed microscopic examination of schistosomula treated with sodium salinomycin demonstrated that the parasites were strongly FDA positive ( with some PI positive cells ) ( Fig . 6C ) , dark and granular ( Fig . 6H and Figure S1 , panel B ) , but motile ( Figure S1 , panel B ) whereas niclosamide treated schistosomula appeared morphologically ( Fig . 6I and Figure S1 , panel H ) and fluorescently ( Fig . 6D ) similar to praziquantel treated parasites ( rounded phenotype , Fig . 6E , Fig . 6J and Figure S1 , panel I ) . Of the six compounds with an unknown , but suggested , anti-schistosome effect , none showed strong decreases in viability as determined by the dual fluorescence assay or epifluorescent microscopy ( e . g . ciclopirox , Fig . 6F ) under the conditions tested in this study . However , upon further microscopic examination , some compounds did induce shape alterations in schistosomula phenotype ( e . g . ciclopirox , Fig . 6K and Figure S1 , panel K ) .
In this study we detail the development and validation of a novel medium- to high- throughput microtiter assay for assessing viability in the schistosomula stage of the human parasite S . mansoni . Infection with this parasitic trematode leads to chronic pathology in endemic populations accounting for disability adjusted life year ( DALYs ) estimates surpassing malaria or tuberculosis [16] , and is responsible for approximately 200 , 000 deaths per annum [2] . With only one drug ( praziquantel ) predominantly used to treat this disease , the potential for drug resistant parasite strains poses a real threat to global control initiatives [17] and supports the vital need for identifying novel therapeutic drug classes or bioactive molecules . To date , high-throughput discovery of new and effective anti-schistosomal compounds has been hampered by the lack of a uniform and quantifiable evaluation method , with current assessment of trematode viability involving subjective and labor-intensive microscopic examination [7] . Application of improved and objective methods in detecting schistosome viability will , therefore , greatly enhance the discovery of next generation chemotherapeutic agents . Here , we provide evidence that a dual-fluorescent , whole schistosomula bioassay ( helminth fluorescent bioassay; HFB ) contains many criteria required for successful high-throughput screening and demonstrate an easily-adaptable methodology by which the Schistosoma genomes [3] , [18] can be rapidly screened for urgently needed drug targets . While many fluorophores are available for use in viability measurements , propidium iodide ( PI ) and fluorescein diacetate ( FDA ) were initially chosen due to their distinct wavelength emission profiles , wide availability from suppliers and low cost . These criteria and the experimental validation that individual schistosomula of polarized viability ( either dead or alive ) only fluoresced red ( PI positive or dead ) or green ( FDA positive or alive ) and not yellow ( Fig . 1 ) , indicated that these two fluorophores stained schistosomula cells independently and could inexpensively be used to quantify parasite viability , when applied to methodology involving a microtiter plate reader equipped to detect fluorescence . While variation in fluorophore ( both PI and FDA ) uptake was noticeable amongst individual schistosomula ( Fig . 1 ) , this was expected due to the non-synchronous nature of schistosome development in vitro as well as experimental factors that commonly affect epifluorescent microscopy including focal point differences between individuals , imperfections in the glass coverslip , localization of a parasite near an air bubble or digital processing of captured images . Importantly , these factors did not interfere with the acquisition of highly-reproducible fluorescence collected from a population of assayed schistosomula when measured by a plate reader ( i . e . Figs . 2–4 ) . Taken together , PI and FDA contained multiple characteristics well-suited for detection of schistosomula viability and , therefore , adapting these fluorophores to microtiter plate formatting was pursued . While minor differences existed between plate formats ( Figs . 2 , 3 and 4A–D; 96- versus 384-well ) , these did not affect the overall quantification of schistosomula viability ( Fig . 4E vs 4F ) . For example , due to smaller surface areas within wells of a 384-well microtiter plate ( compared to 96-well microtiter plates ) , fluorescent intensity slopes ( when correlated to parasite number , Fig . 3 or parasite viability , Fig . 4A–D ) were not as steep when compared to fluorescent intensity measurements collected from 96-well microtiter plates . Nevertheless , strong correlation coefficients , comparing fluorescent intensity units with schistosomula number ( Fig . 3 ) , were obtained in both microtiter plate formats . From these experiments , it was also apparent that FDA could be subjected to spontaneous hydrolysis ( Fig . 2C and 2D ) . While this has been documented before in other systems [15] , and could be due to media components or residual esterase activity released by the dead schistosomes , the fluorescence measured between live and dead parasites were sufficiently large to detect a significant difference ( between 3 min and 12 in both 96-well and 384-well plate formats ) . Nevertheless , substitution fluorophores for FDA , including calcein AM and sulfofluorescein diacetate , are being investigated in current optimizations to the HFB . The observation that calculated , intra-well schistosomula viability ( mathematically derived from plate reader measurements of both PI and FDA , Fig . 4A–D ) paralleled the proportion of viable schistosomula dispensed into each well ( Fig . 4E and 4F ) , demonstrated that FDA and PI could cooperatively be used to sensitively and accurately differentiate amongst percentages of viable parasite populations within either 96- or 384- well microtiter plate formats . This is essential as any high-throughput anthelmintics screening assay may include compounds that do not induce 100% schistosomula lethality and , therefore , use of both PI and FDA allow for accurate quantification across a range of viability endpoints ( between 0% viable to 100% viable ) . Taking advantage of PI and FDA's cooperative ability to differentially stain schistosomula and the capability of a microtiter plate reader to sensitively quantify inter-well parasite viability , we applied the HFB to compounds previously shown to affect schistosomula phenotype and survival [9] , [10] . The first compound selected was auranofin , a potent thioredoxin glutathione reductase inhibitor that has previously been shown to kill larval schistosomula and adult schistosomes [9] . While our HFB confirmed that auranofin does induce schistosomula death ( Fig . 5 ) , our results additionally demonstrated a much finer sensitivity in detecting auranofin LD50 levels ( concentration of auranofin that kills 50% of the assayed biological material ) . Kuntz et al . [9] , using microscopy , reported LD50 levels of auranofin on schistosomula to be between 2 µM and 5 µM . Our auranofin titration series , detected by fluorescent-based microtiter plate quantification , demonstrated that this compound has an anti-schistosomula LD50 activity of 0 . 82 µM±0 . 49 µM . This objective determination is approximately 4 fold more sensitive than what was previously published and clearly demonstrates the increased sensitivity of the HFB and a standardized methodology . Interestingly , schistosomula , treated with 1 µM auranofin ( within the calculated LD50 range ) and visualized by epifluorescent microscopy , displayed cells that were stained with either PI or FDA throughout the body ( Fig . 5G ) . This finding clearly indicates that the HFB is sensitive enough to detect compound-induced changes in fluorescence affecting different cell populations within individual schistosomula as well as between schistosomula and may be useful in identifying compounds that induce cellular stress ( in addition to whole organism viability ) . We next expanded use of the HFB to verify the activity of selected compounds on schistosome phenotype and viability as previously demonstrated by Abdulla et al . [10] . Two of the nine compounds screened showed variance with the published results . The first of these , sodium salinomycin , in contrast to previously published observations [10] , did not induce schistosomula lethality when measured by the HFB or epifluorescent microscopy ( Fig . 6A and 6C ) . The other compound tested that gave conflicting results to that which were published is niclosamide . Niclosamide is recorded in the literature as an anthelmintic , however , its use as such is mainly against cestodes [19] with anti-trematode activity only being reliably recorded against immature Paramphistomum spp . [20] . Its use in schistosomiasis control seems to be limited as a moluscicide for control of the intermediate snail host ( e . g [21] ) . While both sodium salinomycin and niclosamide treated schistosomula were stressed ( decrease in motility - sodium salinomycin , Fig . 6H; shape alterations – niclosamide , Fig . 6I; some PI positive cells – both compounds , Figs . 6C and 6D ) , it is not clear why our results differ from Abdulla et al . [10] in relationship to measured viability . A likely explanation may be associated with how the schistosomula were processed before drug treatment in the two studies . While our study utilized schistosomula , cultured at 37°C , 24 hr after mechanical transformation , Abdulla et al . used schistosomula , maintained on ice , 2 hr after mechanical transformation [10] . We propose that schistosomula , maintained at 4°C , are likely to be more physiologically stressed than parasites cultured at 37°C and , therefore , more susceptible to additional manipulations such as drug treatment . These differences in schistosomula processing , as well as the fact that alternative solvents for niclosamide and sodium salinomycin were used ( ethanol and acetone in this study , whereas DMSO in Abdulla et al . [10] ) , could likely explain the observed differences . The final application of the HFB for measuring schistosomula viability was directed towards the verification of potential novel anthelmintics from those identified by the chemogenomics screening strategy outlined in the recent publication of the S . mansoni genome [3] . Using an iterative bioinformatics approach , Berriman et al . identified 26 S . mansoni target proteins that have strong orthology to human proteins currently being targeted by approved drugs [3] . It was suggested that ‘drug-repositioning’ strategies that re-use existing drugs offer potential savings and cost benefits to the development of novel anthelmintics , especially in the context of neglected tropical diseases like schistosomiasis . Therefore , six drugs with activity against human proteins ( bepridil - calmodulin , ciclopirox - deoxyhypusine synthase , rescinnamine – vesicular amine transporter , flucytosine – bifunctional dihydrofolate reductase-thymidylate synthase , vinblastine – beta tubulin and carbidopa – aromatic amino acid decarboxylase ) were tested against the Schistosoma orthologs ( Smp_026560 , Smp_065120 , Smp_121920 , Smp_135460 , Smp_030730 and Smp_171580 respectively ) using the HFB ( Fig . 6A ) . While none of the selected drugs showed any notable effect on schistosomula viability ( at 10 µM ) , some induced phenotypic changes in parasite morphology ( Fig . 6K and S1 ) and caused a degree of physiological stress ( some PI positive cells , Fig . 6F ) . These compounds , like sodium salinomycin and niclosamide ( see above ) warrant further investigations . The most likely explanation for lack of any detectable anthelmintic activity is that the tested drugs could not efficiently cross the heptalaminate membrane covering the newly transformed schistosomula or that the target proteins were not highly expressed in this larval life stage . Efforts to increase the membrane permeability of all tested drugs ( i . e . standardized solubilization in DMSO ) are currently being implemented in current screens of schistosomula viability , which likely will decrease this potential confounding factor . Recently , both Smp_121920 ( vesicular amine transporter ) and Smp_135460 ( bifunctional dihydrofolate reductase-thymidylate synthase ) were found to be minimally expressed in the schistosomula lifecycle stage as determined by DNA microarray analysis of the parasite lifecycle [22] . These gene expression results could explain the ineffective anti-schistosomula nature of rescinnamine and flucytosine . However , in the same study [22] , Smp_026560 ( calmodulin ) was abundantly found in newly-transformed schistosomula . This finding , supported by proteomic analyses of schistosomula [23] , [24] indicates that calmodulin would be in sufficient quantities for targeting by bepridil . Therefore , although presence of schistosomula targets is necessary for measuring drug effectiveness , it is not sufficient . Other factors ( e . g . drug biotransformation and drug synergy ) are equally important and should be carefully considered during the interpretation of anti-schistosomula drug screening results . In summary , we report a methodology that enables the objective measurement of schistosomula viability , which is quantitative , fast and inexpensive . Although we have restricted its application to drug assays , the adapted methodology has the potential to be used for assessing viability during high-throughput RNAi screens and general manipulations of schistosome development . With this assay functioning in both 96- or 384-well microtiter plate formats , it can be readily adapted for use in academic or industrial settings , in particular if robotics or semi-automation were available . Further refinements of the assay could be proposed to increase processivity and facilitate adaptation to other parasitic and non-parasitic worm species . As well as other worm species , the HFB should also be applicable to other Schistosoma lifecycle stages , including the adult and this enhancement is among other modifications currently being pursued . As death is not the only criterion that could be measured when assessing potential anthelmintic compounds ( compounds that induce stress , paralysis or shape alterations are equally likely to be potent; e . g . niclosamide , sodium salinomycin , ciclopirox , etc . ) , we are currently integrating phenomics , metabolomics and high content screening approaches to add value to our HFB . The combined use of these multiple and quantifiable methods to describe a compound's effect on helminth development will exponentially increase the number of preliminary hits translated into effective and novel anthelmintic chemotherapies . | With only one effective drug , praziquantel , currently used to treat most worldwide cases of schistosomiasis , there exists a pressing need to identify alternative anthelmintics before the development of praziquantel-resistant schistosomes removes our ability to combat this neglected tropical disease . At present , the most widely adopted methodology used to identify promising new anti-schistosome compounds relies on time consuming and subjective microscopic examination of parasite viability in response to in vitro schistosome/compound co-culturing . In our continued effort to identify novel drug and vaccine targets , we detail a dual-fluorescence bioassay that can objectively be used for assessing Schistosoma mansoni schistosomula viability in a medium or high- throughput manner to suit either academic or industrial settings . The described methodology replaces subjectivity with sensitivity and provides an enabling technology useful for rapid in vitro screens of both natural and synthetic compound libraries . It is expected that results obtained from these quantifiable in vitro screens would prioritize the most effective anti-schistosomal compounds for follow-up in vivo experimentation . This highly-adaptable dual-fluorescence bioassay could be integrated with other methods for measuring schistosome phenotype and , together , be used to greatly accelerate our search for novel anthelmintics . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"infectious",
"diseases/helminth",
"infections",
"infectious",
"diseases/neglected",
"tropical",
"diseases"
] | 2010 | Development and Validation of a Quantitative, High-Throughput, Fluorescent-Based Bioassay to Detect Schistosoma Viability |
An important function of all organisms is to ensure that their genetic material remains intact and unaltered through generations . This is an extremely challenging task since the cell's DNA is constantly under assault by endogenous and environmental agents . To protect against this , cells have evolved effective mechanisms to recognize DNA damage , signal its presence , and mediate its repair . While these responses are expected to be highly regulated because they are critical to avoid human diseases , very little is known about the regulation of the expression of genes involved in mediating their effects . The Nucleotide Excision Repair ( NER ) is the major DNA–repair process involved in the recognition and removal of UV-mediated DNA damage . Here we use a combination of in vitro and in vivo assays with an intermittent UV-irradiation protocol to investigate the regulation of key players in the DNA–damage recognition step of NER sub-pathways ( TCR and GGR ) . We show an up-regulation in gene expression of CSA and HR23A , which are involved in TCR and GGR , respectively . Importantly , we show that this occurs through a p53 independent mechanism and that it is coordinated by the stress-responsive transcription factor USF-1 . Furthermore , using a mouse model we show that the loss of USF-1 compromises DNA repair , which suggests that USF-1 plays an important role in maintaining genomic stability .
Maintaining the integrity of the genome through cell generations is critical to ensure accurate cell function and to avoid tumor formation . Cells are continuously challenged by environmental insults and they are equipped with specific and efficient defense machinery to remove any DNA alterations . The importance of these processes is underscored by genetic disorders , such as Bloom , Werner , Cockayne Syndromes and Xeroderma Pigmentosum ( XP ) that result from their impaired function . Despite an enormous amount of progress in identifying the protein complexes and their detailed function in DNA repair pathways , very little is still known about whether these complexes are regulated at a gene expression level . The skin is a good model in which to address this question because it is the organ most exposed to environmental stresses . The principal cause of DNA damage in the skin is solar irradiation , which induces cyclobutane pyrimidine dimers ( CPD ) and 6-4 photoproducts in the epidermal cell layers and which , if not removed , can promote skin cancers . The Nucleotide Excision Repair ( NER ) is the most versatile DNA repair system and is responsible for specifically and constantly eliminating any distorted DNA lesions , including these dimers [1]–[6] . NER can be divided into at least two sub-pathways , Global Genome Repair ( GGR ) [4] and Transcription Coupled Repair ( TCR ) [3] , [5] , [7] . Which one is triggered depends on where the distorted DNA is localized on the genome . GGR , as its name implies , is responsible for removing DNA lesions across the genome including the non-coding part , silent genes and the non-transcribed strands of active genes . The TCR sub-pathway , on the other hand , is dedicated to repairing only DNA lesions detected during transcription and is responsible for removing bulky DNA lesions from the transcribed strands of active genes [2] , [3] . The sequence of events implicated in the GGR and TCR DNA repair pathways include: DNA lesion-recognition ( the rate limiting step ) , DNA-unwinding , excision and repair synthesis and except for the damage recognition step , they share common processes and protein machineries for the remaining events [2] . In the GGR sub-pathway , the XPC-HR23 complex is responsible for the recognition of DNA lesions . The DNA-binding protein , XPC , has a strong affinity for damaged DNA [6] , [8] , [9] . However , its interaction with the evolutionarily conserved HR23 proteins ( homologues of the yeast RAD23 ) is critical for its function . HR23 increases the physiological stability of XPC and thereby its damage recognition activity [10] . In the TCR sub-pathway , lesion recognition occurs through the arrest of the elongating RNA Pol II ( RNAPII ) when it encounters DNA damage . This essential step initiates the subsequent recruitment of the repair factors CSA and CSB , which are required for the removal of the lesion [5] . While it is well accepted that the functional activity of proteins responsible for the removal of DNA-lesions are regulated and indeed crucial to ensure an orchestrated cascade of events [6] , it is not known whether this involves modulation in gene expression . This study addresses this question by using an intermittent UV-irradiation protocol and investigates the gene expression profile of key players in the NER DNA-damage recognition step . We show that UV-induced DNA photo-lesions initiate a specific program of gene expression with the stress responsive transcription factor Upstream Stimulatory Factor 1 ( USF-1 ) playing a central role [11]–[13] . Using a combination of in vivo and in vitro assays we demonstrate , in our system , that there is a specific and coordinated regulation of HR23A , HR23B , CSA and CSB genes and their protein levels in response to UV-mediated DNA damage . We show that up-regulation of both HR23A and CSA is driven by a common p53 independent mechanism involving USF-1 . Furthermore , we provide novel evidence that while HR23A and HR23B share a similar function in DNA-damage recognition , their temporal expressions are different , which may imply that they function at different times , in response to UV-induced DNA-damage . Results from this study have important implications for our understanding of the role of gene expression regulation in the DNA-damage repair pathways and reveal a role for USF-1 in DNA-repair and in maintaining genome integrity .
Very little is known about how genes that encode key components of the NER recognition step are regulated at a transcriptional level , to mediate their role in DNA lesion recognition . We thus performed a UV-induced DNA-lesion protocol ( Figure 1A ) , which generates immediate DNA photo-lesions through repetitive doses of short wavelength UV pulses rather than delivery of a single high dose [14] , [15] . Using RT-qPCR , we then followed the expression of genes specifically involved in the recognition events of TCR ( CSA and CSB ) and GGR ( HR23A and HR23B ) , immediately post-irradiation . Cultured mouse and human keratinocytes ( XB2 , HaCaT ) were irradiated with four to eight 10 J/m2 UV pulses ( 254 nm ) at 15 min intervals and collected at the indicated times ( from 30 min to 5 h ) after the last pulse ( Figure 1A ) . We first checked for the presence of CPD post UV-irradiation ( Figure S1 ) and for cell viability over 24 h confirming that the irradiation procedure was inducing DNA-damage without compromising cell numbers ( 90% and 75% cell survival at respectively 3 and 24 h ) ( Figure 1B ) . The irradiation protocol resulted in a significant increase of CSA mRNA levels ( 6-fold after 30 min ) , while the abundance of CSB gene transcripts was not affected ( Figure 2A ) . CSA mRNA levels remained elevated at 1 h and decreased from 2 hours . Comparable results were obtained in p53-deficient human HaCaT keratinocytes [16] ( Figure S1A ) . Up-regulation of CSA gene expression was accompanied by a significant increase in CSA protein levels ( Figure 2B ) , peaking at 3 hours compared to un-stimulated cells , where CSA protein is almost undetectable . The increase of CSA protein levels following UV-irradiation was also observed by immunofluoro-staining in XB2 keratinocytes ( Figure S1B ) . This increase in CSA protein levels is significantly reduced over time when cells were pre-treated with α-amanitin , an agent that disrupts transcription . These results indicate that the increase in protein levels results in part from transcriptional regulation . We next investigated the regulation of the GGR pathway-specific mediators , HR23A and its homologue HR23B , following the same irradiation protocol ( Figure 1A ) . While no significant effect was observed on HR23B mRNA levels , the irradiation protocol resulted in a mild but reproducible ( 6 independent experiments ) and significant increase of HR23A mRNA levels ( 1 . 5-fold at 4 h ) ( Figure 2C ) . Comparable results were obtained in p53-deficient human HaCaT keratinocytes ( Figure S1C ) . In parallel with UV-induced HR23A transcripts , protein levels increased progressively , reaching a 4-fold increase at 8 hours . This effect was abrogated when cells were pre-treated with α-amanitin ( Figure 2D ) . The increase of HR23A protein levels following UV-irradiation was also observed by immunofluoro-staining in XB2 keratinocytes and correlates with an increase in CPD ( Figure S1D ) . In contrast , HR23B protein levels decreased over time after UV-irradiation suggesting that it is regulated post-transcriptionally since there was no change in its mRNA levels ( Figure 2C ) . These results indicate that HR23A and HR23B are regulated differently . UV-induced transcription is a tightly regulated process that involves both cis and trans UV-responsive elements . We thus explored potential cis/trans factors involved in UV-induced regulation of CSA and HR23A expression by in silico analysis of their respective proximal promoter sequences using Consite and Zpicture ( Rvista 2 ) softwares [17] , [18] . We found that both promoters belong to the TATA-less class and that their proximal regions contain consensus E-box motifs ( CACGTG ) upstream from the transcription start site ( TSS ) at −246 for CSA ( Figure 3A ) and −154 and −37 for HR23A promoter ( Figure 3D ) , which are highly conserved across species . By contrast , no such conserved E-box motif was found in the CSB and HR23B promoter regions ( data not shown and Figure 3D ) . Given that USF-1 acts as a key player of UV-regulated gene expression by interacting specifically with E-box cis-regulatory elements ( CACGTG ) as homodimers or as heterodimers with its partner USF-2 [19]–[21] , we suspected that CSA and HR23A may be USF-1 target genes . To test this hypothesis , we performed chromatin immunoprecipitation ( ChIP ) assays using antibodies specific for either USF-1 or USF-2 . DNA recovered from the HaCaT cell line was amplified by PCR using primers targeting distinct promoter sequences ( Figure 3B ) . Results showed specific amplification products corresponding to the binding of USF-1 and USF-2 factors to the CSA proximal promoter ( −246 bp ) , whereas no PCR product was observed for the distal region ( −2 kb ) , the proximal CSB promoter or with non-specific IgG antibodies ( Figure 3B ) . We next investigated the impact of UV-mediated DNA-damage ( 8×10 J/m2 ) on the recruitment of USFs to the CSA proximal promoter over time . UV-irradiation specifically and rapidly ( 15 min ) promoted an 8-fold enrichment of USF-1 , but not USF-2 , at the CSA proximal promoter ( Figure 3B ) . Using in vitro binding assays ( EMSA ) , we tested the ability of USFs to bind the identified conserved E-box motif ( −246 bp ) , which was also present in the ChIP amplified product . Specific DNA-protein complexes were obtained with a probe spanning the E-box motif at −246 ( Figure 3C ) , which were efficiently competed by homologous cold wild type , but not mutant probe . These DNA-protein complexes were super-shifted by antibodies against either USF-1 or USF-2 but not by non-specific antibodies ( IgG or Tbx2 ) . No DNA-protein complex was formed with probes carrying mutated E-box sequences . In vivo DNA-binding assays revealed also that USF factors interact specifically with the HR23A proximal promoter but not the distal promoter or HR23B promoter and that UV-irradiation promotes the interaction of the USF-1 transcription factor by a 3-fold and USF-2 by a 2 . 5-fold enrichment ( Figure 3E ) . As shown previously for CSA , EMSA assays confirmed the DNA-protein complexes spanning the conserved −154 and −36 E-box motifs , present also in the ChIP amplified products ( Figure 3F ) . Competition between the two E-box probes did not reveal any preferential binding site ( data not shown ) . In addition to the E-box sites present in the HR23A proximal promoter , in silico analysis identified conserved GC-rich regions ( −131 and −18 from TSS ) ( Figure 3D ) known to interact with members of the SP1/SP3 transcription factor family [11] . To examine their respective contribution to the regulation of HR23A expression , we performed in vitro and in vivo DNA-binding assays as described above . Specific protein-DNA complexes were formed only in the presence of the −131 intact GC box that interacts with SP1 and SP3 transcription factors ( Figure 3G ) . Also , under the experimental conditions used , only SP3 was able to bind the HR23A proximal promoter in vivo and SP3 loading was not affected by UV-irradiation . Interestingly , a comparable SP3 binding profile was obtained with the HR23B proximal promoter that shares homologous GC rich sequences with HR23A ( Figure 3H ) but whose mRNA levels were not modulated by UV , suggesting that the GC motifs might not be UV-inducible . Taken together these results provided compelling evidence that , in response to UV-irradiation , USF-1 interacts directly with the CSA and HR23A proximal promoters , suggesting it may be responsible for the UV-induced CSA and HR23A expression observed in this study . The relevance of the E-box motifs in mediating USF regulation of the CSA and HR23A promoters was next assessed by luciferase assays . We first transiently co-transfected XB2 cells with a wild type ( WT ) and E-box mutated CSA promoter ( −847/+1 ) cloned upstream of a luciferase reporter ( pGL3-Luc ) ( Figure 4A ) and USF-1 or USF-2 expression vectors ( pCMV ) [12] , [20] . Both USF-1 and USF-2 expression vectors led to significant increases of CSA-luciferase activity ( Figure 4B ) . Following UV-irradiation , WT CSA promoter activity demonstrated a rapid , 6-fold significant increase ( 30 min after the last UV pulse ) ( Figure 4C ) . Furthermore , this intact E-box cis-regulatory element proved to be required for UV-induced activation and to mediate the binding of USF trans-activators ( Figure 4B–4C ) . We next transiently co-transfected XB2 cells with a WT and E-box mutated HR23A promoter ( −186/+73 ) construct cloned upstream of a luciferase reporter ( Figure 4D ) . USF-1 and USF-2 expression vectors led to mild but significant increases of HR23A-luciferase activity ( Figure 4E ) . In response to UV-irradiation , HR23A promoter activity increased slightly but significantly only in the presence of the USF-1 expressing vector ( Figure 4G ) . Interestingly , when the two E-box motifs were mutated , we observed a 4-fold reduction of the basal HR23A-luciferase activity ( Figure 4F ) and the USF-1 mediated UV-response was abrogated ( Figure 4G ) . Mutation of the −131 GC-rich motif did not significantly affect HR23A basal activity and did not impair the USF-mediated UV-response ( Figure 4F–4G ) , supporting the idea that the UV response is driven by the USF/E-box protein/DNA complexes . The physiological significance of the regulation of CSA and HR23A by USF-1 in response to UV-induced DNA damage was established using genetic approaches with XB2 USF-1 knock-down ( KD ) cells ( Figure 5 ) and USF-1 knock-out ( KO ) mice ( Figure 6 ) [13] . Firstly , we quantified the level of CPDs in cells in which either USF-1 or CSA or HR23A mRNA was targeted with two different and independent siRNA . While the level of CPDs in the un-stimulated USF-1-KD cells ( siUSF-1 N°1 and N°2 ) remained low and comparable to the control cells ( siCtrl N°1 and N°2 ) , the level increased dramatically 4 hours following UV exposure and was significantly higher than the control cells exposed to UV . Surprisingly , although confirmed by two independent siRNAs , levels of CPDs in CSA-KD cells ( siCSA N°1 and N°2 ) and in HR23A-KD cells ( siHR23A N°1 and 2 ) were both significantly elevated in the absence of UV-irradiation compared to USF-1-KD and control cells . Following UV-irradiation , there was a mild increase in levels of CPDs in CSA-KD and HR23A-KD cells which was probably due to the initial high level of CPDs in KD-cells coupled to quantification limits . Nonetheless , these increases remained significantly higher compared to irradiated control cells ( Figure 5 ) . Secondly , using skin punch biopsies prepared from USF-1 KO mice and the WT littermates , we analyzed the UV-response by comparing gene transcription efficiency and levels of CPD . RNA analysis comparing irradiated versus non-irradiated WT skin punch biopsies showed that CSA and HR23A mRNA increased 3 . 5-fold and 2 . 5-fold at 1 and 5 h post-irradiation , respectively ( Figure 6A–6B ) . CSA and HR23A transcript levels remained at basal levels in USF-1 KO mice and CSB and HR23B mRNA were not affected by UV-irradiation in both WT and KO mice ( Figure 6A–6B ) . By contrast , UV-inducible but USF-1-independent genes , such as the Gadd45α prototype displayed UV-induced transcript profiles in WT and KO USF-1 mice ( Figure 6C ) [22] , [23] . However , we detected a 2 h delay of the mRNA increase in the USF-1 KO mice ( Figure 6C ) , which is consistent with RNAPII being arrested to permit DNA-repair of transcribed genes before the commencement of transcription supporting that TCR is compromised in USF-1 KO mice . Moreover , because HR23 proteins are crucial to stabilize XPC at the DNA-photolesion sites to permit removal of damage , we quantified the level of CPD by ELISA immediately after 3 UV-pulses , and after 4 UV-pulses over 36 h ( Figure 6D ) . While basal levels of CPD were comparable in both WT and USF-1 KO mice as expected from siRNA results , UV-irradiation led to rapid increases of DNA-damage that were comparable immediately after 3 UV-pulses but remained higher over time in KO mice compared to WT mice after 4 UV-pulses . Importantly , calculating the rate of CPD-clearance over 36 h , we observed a difference between WT and KO mice ( Figure 6D ) . Whereas CPDs were removed in WT mice at 36 h , CPDs remained elevated in USF-1-KO mice at this time point . Taken together these results provide compelling evidence that in response to UV-induced DNA-damage , loss of USF-1 compromises the tight regulation of the NER resulting in altered removal of UV-induced DNA-damage .
DNA carries the genetic instruction required for the development and functioning of all living organisms . This information must be transmitted to daughter cells with high fidelity , and therefore specific DNA-repair programs are present to eliminate DNA-lesions produced by regular threats . The NER pathway is dedicated to repair distorted DNA , and for decades studies have focused on elucidating the molecular mechanisms involved in the recognition , signaling and removal of these DNA-lesions [2] , [24] . Using a multiple dose UV-irradiation protocol with repetitive lower UV-doses that more accurately mimics our exposure to solar irradiation compared to a single high dose , our study identifies an early and coordinated gene expression regulation program of the CSA and HR23A genes in mammals that relies on the presence of the USF-1 transcription factor . CSA and CSB proteins have been shown to have dedicated and specific functions in the TCR pathway [5] . It has indeed been clearly established , even in the absence of DNA damage , that a large part of the CSB protein is found associated with chromatin and that RNAPII even in the absence of DNA damage , and this association increases upon UV-irradiation [25] , [26] . CSA has however been shown to interact indirectly with RNAPII [25] , but it is required in cooperation with CSB for the recruitment of XAB2 , HMGN1 and TFIIS , to trigger DNA repair mediated by XP complexes and PCNA protein [5] , [24] . The importance of CSA in the early DNA damage response might also reside in the timing of its specific gene expression as its levels are low in resting cells but increase dramatically immediately after UV-irradiation . One possible explanation would be that because CSA acts as a unique player in the initial step of TCR , appropriate levels of the protein is required almost immediately after UV-induced DNA damage and before RNAPII gets arrested by de novo DNA photo-lesions . No increase in CSA protein leads to a delay in transcription likely by an impairment of its associated function: recruitment and stabilization of the initiation complex on the chromatin [25] . This is also supported by deficient CSA being directly linked to the Cockayne syndrome type A genetic disorder [27] and by siRNA results . However , these patients are not prone to developing skin-cancers like XP patients , presumably due to 1- the presence of additional DNA-repair machinery operating post DNA-replication [28] , 2- increased cell-death after DNA-damage [28] and to an average life-span for these patients generally being limited to 12 years [29] . Interestingly , specific mutations in the repair-enzyme genes XPB , D and G produce phenotype reflecting a combination of traits present with XP and CS syndromes . This suggests that simultaneous alteration of GGR and TCR will promote mutagenesis in certain cells [29] . HR23A and HR23B proteins share common domains and are both able to form a complex with XPC [30] , [31] . The XPC-HR23B complexes were however reported to be more abundant than the XPC-HR23A complexes and have been shown to participate almost exclusively in DNA-photolesions recognition in vivo [32] . The XPC-HR23A complexes were consequently regarded as having a functionally redundant role to XPC-HR23B . This is therefore the first study to report conditions under which HR23A and B protein levels are modulated differently , which suggest that HR23A may have a function distinct from HR23B in the UV-induced DNA damage pathway . We show that in response to repetitive UV-irradiation there is a 4-fold increase in the level of HR23A protein which is associated with a concomitant loss of HR23B and we propose that this may favor XPC-HR23A complex formation which leads to sustained XPC-stabilization for appropriate recognition of DNA lesions [32] , [33] . Indeed , while HR23A and HR23B KO mice are NER proficient , double HR23A and HR23B KO derived cells show an XPC-like phenotype [34] . We propose that differential regulation of these two HR23 homologues may provide a safety mechanism to ensure the stability of XPC and its function in response to multiple UV-exposure . This possibility is supported by our data that show ( i ) a reduction of DNA–lesion removal in HR23A-KD cells , ( ii ) a reduction of DNA-lesion removal in UV-irradiated USF-1 KO tissue and KD cells , which occurs presumably in part due to an abrogation of HR23A gene expression in response to UV-rays and ( iii ) a diminution of HR23A protein when UV-induced HR23A transcription is abrogated with α-amanitin . We thus believe that our study reveals a difference in the DNA damage response to a single high dose of UV-irradiation compared to repetitive lower doses and that our conditions mimic the accumulation of DNA-damage over a short period of time which is more applicable to every day life . These results are particularly interesting in the light of the Saccharomyces cerevisiae RAD23 gene , the ortholog of both HR23A and HR23B , which also presents with an UV-inducible phenotype [35] . Our results show that the UV-induced function has been conserved through evolution and restricted to one member for specific regulatory purposes . USF-1 is activated by the stress-dependent p38 kinase and then operates as a transcriptional rheostat of the stress response [20] , [21] . Combined regulation of HR23A and CSA gene expression by USF-1 thus allows a tight and sequential regulation of these two genes . The observation that there is first an increase in USF-1 occupancy on the CSA promoter followed by its occupancy on the HR23A promoter suggests a sequential and dynamic recruitment of USF-1 to fulfill specific steps of a common task . USF-1 as a stress responsive factor is also proposed to be a key player in regulating pigmentation gene expression in response to UV-irradiation [12] , [21] , [36] . USF-1 may thus elicit a skin protection program against UV-induced DNA damage by controlling two independent and complementary pathways: the DNA-photolesions repair process and the UV-induced tanning response . More importantly , USF-1 functions independently of p53 but both pathways are expected to be coupled [37] . Since USF-1 mediates an independent and crucial DNA-repair program as highlighted by our USF-1 KO and KD assays , we propose that impairment of this pathway will promote genome instability in response to environmental insults , which is a hallmark of cancer . This hypothesis is supported by the reported loss of USF activity in breast cancer cells [38] , and impairment of the recruitment of USF factors to specific E-box elements due to SNPs , as observed in the variant rs1867277 FOXE1 gene , conferring thyroid cancer susceptibility [39] . Furthermore , CpG methylation can also impair USF interaction with core E-box motifs and subsequently alter gene expression , as for the metallothionein-I gene which is silenced in mouse lymphosarcoma [40] . Our findings indicate that , in response to repetitive environmental threats that lead to the accumulation of UV-induced DNA damage , the NER pathway undergoes a program of gene expression that correlates with the DNA repair processes and that the USF-1 transcription factor is central to this program . These results may thus have important implications for our global understanding of how genome instability is promoted .
HaCaT ( human - p53 deficient ) and XB2 ( mouse ) keratinocytes were maintained in D-MEM ( Invitrogen ) supplemented with 10% FBS ( Sigma ) and 1% Penicillin-Streptomycin ( Invitrogen ) at 37°C in 5% CO2 atmosphere . Skin biopsies ( 0 . 8 cm diameter ) were recovered from the backs of WT and USF-1 knockout mice ( 8 weeks ) [13] and maintained in culture for up to 24 h in RPMI ( Invitrogen ) supplemented with 1% Penicillin-Streptomycin at 37°C in a 5% CO2 atmosphere . Specific DNA photo-lesions were generated with ultraviolet bulbs ( 254 nm ) [14] , using the Stratalinker apparatus ( Stratagene ) as previously described [12] , [20] , [21] . The day before UV exposure , cells were plated at 50–70% confluence , depending on their doubling time , in 10 cm Petri dishes . Twelve to twenty-four later , the medium was replaced with fresh medium supplemented with 2% FBS and 1% antibiotics . The following day , cells were UV irradiated ( 2× to 8× 10 J/m2 ) . UV pulse set at 10 J/m2 lasted 3 seconds . The medium was completely removed before and replaced after irradiation . At the time point indicated , cells were washed twice in cold PBS , harvested by scraping , centrifuged and resuspended in appropriate buffer . For transcription inhibition experiments , cells were pre-treated with α-amanitin ( 5 µg/ml; Sigma ) 30 min prior to UV-irradiation . Mouse skin biopsies were irradiated with four successive pulses of 50 J/m2 UV , recovered at the indicated time points by placing the skin biopsy directly in RNA later buffer ( Qiagen ) and stored at −20°C for subsequent RNA extraction . Cell viability in response to UV ( 254 nm ) was analysed in 96 well plates . Briefly , cells were plated at 1×104 cells/well , 10 h before UV induction , tetrazolium salt ( MTT , 0 . 5 mg/ml ( Sigma ) was added to culture medium . After 3 h of incubation ( 37°C ) , the medium was removed and 150 µl of DMSO was added to each well . Percentage of cell viability was then analysed by measuring the DMSO-optical density ( OD ) , at 690 and 540 nm with a Multiskan spectrophotometer . RNA was extracted using Nucleospin RNA II kit ( Macherey Nagel ) and quantified using the Nanodrop device . For skin explants , an extra Trizol/chloroform purification step was needed to remove protein . cDNA was obtained by reverse transcription using a High-Capacity cDNA Reverse Transcription Kit ( Applied Biosystem ) from 1 µg of total RNA . Gene expression was analyzed by qPCR in sealed 384-well microtiter plates using the SYBR Green TM PCR Master Mix ( Applied Biosystem ) with the 7900HT Fast Real-Time PCR System ( Applied Biosystem ) . Relative amounts of transcripts were determined using the delta Ct method . The mRNA levels at each time point following stimulation are expressed as fold increase , relative to non-irradiated cells . Data were normalized independently to at least two housekeeping genes HPRT and GAPDH . Because comparable data were obtained only the HPRT ones are presented . Each experiment was carried out at least twice and each time point was repeated in triplicate . Forward ( F ) and reverse ( R ) primers were designed using the Universal Probe Library Assay Design Center ( Roche ) and have been previously tested for their efficiency ( Sequences available on request ) . Harvested cells were immediately lysed by incubation for 30 min in ice-cold RIPA buffer ( supplied in protease and phosphatase inhibitors ) . Equal amounts of protein were denaturated in Laemmli buffer for 5 min at 95°C and resolved by 15% SDS-PAGE . Membranes were probed with appropriate antibodies and signals detected using the LAS-3000 Imaging System ( Fujifilm ) were quantified with ImageJ ( http://rsbweb . nih . gov/ij/ ) . Gel electrophoresis DNA binding assays were performed with crude HaCaT keratinocyte nuclear extracts under conditions previously described [12] , [41] , [42] , with modifications . Double-stranded oligonucleotides were labeled with T4 polynucleotide kinase in the presence of P32-γATP ( 3000 Ci/mmol ) and purified in columns ( Mini Quick Spin Oligo Columns , Roche Diagnostic ) . Reaction mixtures contained 2–4 µg of total protein and 0 . 03 pmol of P32 end-labelled probe in binding buffer ( Hepes 25 mM , KCl 150 mM , 10% Glycerol , DTT 10 mM , 1 µg of poly ( dIdC ) , 1 µg salmon sperm DNA ) . After 20 min of incubation , samples were loaded onto a low ionic strength 6% polyacrylamide gel ( 29∶1 cross-linking ratio ) containing Tris Borate Na EDTA buffer pH 8 . 3 . Supershift and competition assays were performed by adding competitor probes ( 1× to 100× ) or antibodies ( 0 . 2 µg ) prior to incubation with labelled probes ( Sequences available on request ) . Radioactive bands were quantified with a STORM 840 PhosphorImager ( Molecular Dynamics ) . ChIP assays , using 1 . 5–2×106 HaCaT cells , were performed as previously described [41] , [43] , with specific adaptations . The cells were cross-linked ( 1 . 5% formaldehyde ) , washed twice and collected in 1 ml cold PBS . Cells were lysed and the samples were then sonicated for DNA fragmentation ( Sonifier Cell Disruptor , Branson ) in 1 ml lysis buffer ( 10 mM EDTA , 50 mM Tris-HCl ( pH 8 . 0 ) , 1% SDS , 0 . 5% Empigen BB ) and diluted 2 . 5-fold in IP buffer ( 2 mM EDTA , 100 mM NaCl , 20 mM Tris-HCl ( pH 8 . 1 ) , 0 . 5% Triton X-100 ) . This fraction was subjected to immunoprecipitation overnight with 3 µg of the appropriate antibody . These samples were then incubated for 3 h at 4°C with 50 µl of protein A-Sepharose beads slurry . Precipitates were washed several times , cross-linking reversed and DNA purified using a Nucleospin Extract II kit ( Macherey Nagel ) . PCR or qPCR analyses were carried out with primers spanning HR23A , HR23B and CSA proximal promoters or , as a reference , with primers targeting an unrelated promoter region ( HSP70 promoter region ) or unspecific regions of target promoter genes ( sequences available on request ) . End-point PCR was performed in semi-quantitative conditions for ChIP ( 30 amplification Cycles ) . For qPCR analysis , fold enrichment was determined using the ΔΔCt method: Fold enrichment = 2− ( Δct1−ΔCt2 ) , where ΔCt 1 is the ChIP of interest and ΔCt2 the control ChIP . −744/+73 and −185/+73 HR23A promoter region were obtained by PCR and inserted into the luciferase reporter plasmid pGL3-basic ( Promega ) . E boxes and the GC box were mutated using a QuickChange Site-Directed Mutagenesis Kit ( Stratagene ) . The same protocol was used for the CSA promoter sequence lying −847/+1 . CSA and HR23A promoter regulation was studied in mouse XB2 keratinocytes . Cells were plated at 60–70% confluence in 12-well plates in medium supplemented with 10% SVF without antibiotics and were maintained for 12 h . Cells were co-transfected or not with pGL3 reporter vector and pCMV ( empty , USF-1 or USF-2 ) , as previously described [12] , [20] , [21] . The transfection mix , containing up to 500 ng of plasmid DNA , was prepared in Optimem medium ( Invitrogen ) and used to transfect cells for 3 h using Lipofectamin 2000 ( Invitrogen ) . 3 h after transfection , the medium was replaced with fresh medium supplemented with 10% SVF and 1% antibiotics . 48 h later , cells were irradiated with UV , as described above and harvested up to 5 h following UV . Cells were then passively lysed and luciferase activity was quantified in a Microplate Luminometer Centro LB 960 ( Berthold ) using the Luciferase Reporter Assay System ( Promega ) . XB2 cells were seeded and transfected in 10 cm-diameter dishes ( 1×106 cells per dish ) in DMEM medium complemented with 10% FBS , with 40 pmol of siRNA . Two different siRNA ( N°1 and 2 ) were used independently for each target gene tested ( CSA , HR23A and USF-1 ) as for control ( siOTP1 , siNT1 ) ( Sigma-Genosys , St Louis , MO ) using Lipofectamine 2000 ( Invitrogen , Paisley , UK ) . Transfections were performed following provider's instructions . 72 hours later , the cells were UV irradiated as previously described and recovered 4 hours after the irradiation protocol for CPD quantification and western blot analysis . siRNA sequences are available on request . Quantifications of CPD in skin explants following UV ( 254 nm ) ( 4×50 J/m2 ) were performed by ELISA , accordingly to Cosmo bio recommendations . DNA purification was performed by phenol/chloroform extraction and ethanol precipitation . Briefly , 200 ng of denatured DNA was distributed onto protamine sulfate precoated 96 well plates ( Polyvinylchloride flat-bottom ) . Detection of DNA-lesion was performed using specific mouse anti-CPD antibodies , and revealed with the biotin/peroxidase-streptovidin assay . Quantification was obtained by the absorbance at 492 nm . Each experiment was performed independently with punch biopsies of three independent WT and USF-1 KO mice . XB2 ( mouse keratinocytes ) cell lines were cultured in D-MEM at 37°C on glass coverslips in 35-mm dishes . 24 hours later , cells were UV-irradiated with 6×10 J/m2 in serum free medium following as previously described . Cells were then fixed and permeabilized after different times of induction accordingly to Cosmo bio Co protocol . Previously to CPD immunostaining in cells , we denatured DNA with HCl 2 M for 30 min at room temperature . Indirect immunofluorescence was then performed using specific recommendations of Cosmo bio Co protocol with specific primary antibodies mouse anti-CPD ( TDM2 clone , MBL ) ( 1∶3000 ) . Fluoro-staining was performed with labeled donkey anti-mouse IgG ( Alexa Fluor 488 ) . CSA immunostaining was performed with specific anti-rabbit antibody from Santa Cruz . Anti USF-1 ( C:20 ) , USF-2 ( N-18 ) , Sp1 ( PEP 2 ) , Sp3 ( D-20 ) , TBX-2 ( C-17 ) , HR23B ( P-18 ) , HSC70 ( B-6 ) were purchased from Santa Cruz . Anti HR23A ( ARP42211 ) was purchased from Aviva . Anti CSA was purchased from Abcam ( ab96780 ) . Anti CPD ( TDM2 ) was purchased from MBL . Anti α-Tubulin ( ARP42211 ) was purchased from Sigma . Errors bars represent standard deviation , stars indicate statistically significant differences ( two-tailed Student's t-test ) between control and irradiated samples * P<0 . 05; ** P<0 . 01; *** P<0 . 001 . The present animal study follows the 3R legislation ( Replace-Reduce-Refine ) . It has been declared and approved by the French Government Board . Animal welfare is a constant priority: animals were thus sacrificed under anesthesia . | UV is responsible for DNA damage and genetic alterations of key players of the Nucleotide Excision Repair ( NER ) machinery promote the development of UV-induced skin cancers . The NER is the major DNA–repair process involved in the recognition and removal of UV-mediated DNA damage . Different factors participating in this DNA repair are essential , and their mutations are associated with severe genetic diseases such as Cockayne Syndrome and Xeroderma Pigmentosum . Here , we show for the first time that the specific regulation of expression in response to UV of two NER factors CSA and HR23A is required to efficiently remove DNA lesions and to maintain genomic stability . We also implicate the USF-1 transcription factor in the regulation of the expression of these factors using in vitro and in vivo models . This finding is particularly important because UV is the major cause of skin cancers and dramatically compromises patients with highly sensitive genetic diseases . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"dermatology",
"biology",
"molecular",
"cell",
"biology"
] | 2012 | USF-1 Is Critical for Maintaining Genome Integrity in Response to UV-Induced DNA Photolesions |
Genome-wide association studies ( GWAS ) have identified many genetic susceptibility loci for colorectal cancer ( CRC ) . However , variants in these loci explain only a small proportion of familial aggregation , and there are likely additional variants that are associated with CRC susceptibility . Genome-wide studies of gene-environment interactions may identify variants that are not detected in GWAS of marginal gene effects . To study this , we conducted a genome-wide analysis for interaction between genetic variants and alcohol consumption and cigarette smoking using data from the Colon Cancer Family Registry ( CCFR ) and the Genetics and Epidemiology of Colorectal Cancer Consortium ( GECCO ) . Interactions were tested using logistic regression . We identified interaction between CRC risk and alcohol consumption and variants in the 9q22 . 32/HIATL1 ( Pinteraction = 1 . 76×10−8; permuted p-value 3 . 51x10-8 ) region . Compared to non-/occasional drinking light to moderate alcohol consumption was associated with a lower risk of colorectal cancer among individuals with rs9409565 CT genotype ( OR , 0 . 82 [95% CI , 0 . 74–0 . 91]; P = 2 . 1×10−4 ) and TT genotypes ( OR , 0 . 62 [95% CI , 0 . 51–0 . 75]; P = 1 . 3×10−6 ) but not associated among those with the CC genotype ( p = 0 . 059 ) . No genome-wide statistically significant interactions were observed for smoking . If replicated our suggestive finding of a genome-wide significant interaction between genetic variants and alcohol consumption might contribute to understanding colorectal cancer etiology and identifying subpopulations with differential susceptibility to the effect of alcohol on CRC risk .
Colorectal cancer ( CRC ) is the third-most common cancer in men and the second most common cancer in women worldwide [1] . Both environmental and genetic factors are involved in the development of CRC [2–7] . Since 2007 , genome-wide association studies ( GWAS ) have identified about 50 loci associated with CRC risk[8–11] . However , only a small portion of the familial aggregation of CRC is explained by these identified genetic loci , and additional variants associated with CRC susceptibility are more likely to be identified through analyses of interactions between genes and environmental risk factors [12 , 13] . Single nucleotide polymorphisms ( SNP ) that impact only a subgroup of the population or have opposite effects in different subgroups are likely to produce weak main effects that cannot be easily detected by marginal association testing of the SNPs . However , these variants may be identified by testing for interactions between SNP and environmental risk factors ( genome-wide interaction analysis ) [14 , 15] . These findings may provide etiologic insight into CRC and identify potentially susceptible subpopulations [14 , 15] . There is compelling evidence from epidemiologic studies that alcohol consumption and cigarette smoking are associated with risk of CRC [16–25] . Both alcohol consumption and cigarette smoking influence disease risk through pathways involving multiple gene products and regulatory elements , providing potential for biological interactions [26–28] . Accordingly , alcohol consumption and smoking are important lifestyle factors to study interactions with genetic variants . In this study , we performed a genome-wide interaction analysis using the large datasets from the Colon Cancer Family Registry ( CCFR ) and the Genetics and Epidemiology of Colorectal Cancer Consortium ( GECCO ) [3] to identify SNPs that modify the effects of alcohol and smoking on CRC risk .
The SNP rs9409565 showing a significant interaction with alcohol is located in an intergenic region between HIATL1 and FBP2 . As there is a recombination hotspot lying between rs9409565 and FPB2 ( Fig 3 ) , we focused the gene expression analysis on HIATL1 , which is expressed in normal colon and rectal tissue . [34 , 35] Furthermore , based on our gene expression data for 35 colorectal cancer cases ( S2 Text ) , the expression levels of the HIATL1 gene was significantly higher in tumor tissues compared with adjacent normal tissues ( paired student t test , P<7 . 2×10−5 , S2 Fig ) . This finding is consistent with a previous study [36] which is included in the UCSC Cancer Genomics Browser[37–39] and show that human colon tumors ( n = 100 ) significantly over-expressed HIATL1 compared to normal colon tissues ( n = 5 ) [36] ( Fisher exact test: P = 0 . 03 ) . Similarly , we were able to reproduce this observation in 50 independent paired colorectal adenocarcinoma and adjacent normal samples from The Cancer Genome Atlas ( TCGA ) ( paired student t test , P = 0 . 02 , S2 Fig ) . Furthermore , we observed that HIATL1 showed significant differential expression across various levels of lifetime alcohol consumption in the colon tumor tissues ( n = 28 , ANOVA test P = 0 . 03 , S3 Fig ) and also had differential gene expression across levels of alcohol consumption at reference time ( the year before enrollment ) in the normal colon tissues ( n = 33 ) at P = 0 . 06 from ANOVA test ( S4 Fig ) . In addition , for rs9409565 and rs9409567 ( LD r2 = 1 . 0 in CEU population ) , the two most significant SNPs at 9q22 . 32/HIATL1 , are cis-acting quantitative trait loci ( eQTL ) for HIATL1 expression in lymphoblastoid cell lines ( P<7 . 0×10−6 ) and monocytes ( P<5 . 8×10−12 ) [40 , 41] , which is consistent with previously published eQTL results from GTEx , Genevar[42] , Westra et al . , and Lappalainen et al . showing that this these SNPs tag an eQTL locus in lymphoblastoid cells and related anatomical sources ( including spleen , whole blood , esophagus muscularis , and sun-exposed skin ) with p values ranging from 7x10-138 to 4x10-6 ( S8 Table ) . In contrast , evaluation of eQTL in both normal ( GTEx ) and cancer colorectal tissue from TCGA for the rs9409565 locus ( r2> = 0 . 2 in Phase 3 1000 genomes EUR data ) did not show any significant eQTL . The inability to detect an eQTL is likely because the enhancer tagged by the locus is active in some but not all cancer cell lines and the current reference cancer transcriptome data may not be large enough or molecularly representative of our study population S5 Fig ) . Furthermore , we investigated whether any of the tagging SNPs are located in variant enhancer loci ( VEL ) reported by Akhtar-Zaidi et al . [43] using ChIP-seq ( H3k27ac ) enhancer signals . We observed that four of the variants ( rs28406858 , rs7042481 , rs7858082 , and rs9409510 ) in LD with rs9409565 ( LD r2≥0 . 6 ) were positioned within three gained cancer-specific VEL ( S6 Fig ) .
We identified a suggestive interaction between variants at 9q22 . 32/HIATL1 and light-to-moderate alcohol consumption in relation to CRC risk . This is the first genome-wide significant GxE interaction reported for alcohol intake and risk of CRC and warrants replication in independent studies . Evidence for overlap between the discovered 9q22 . 32/HIATL1 region with VEL as well as gene expression results support the relevance of the 9q22 . 32/HIATL1 region for CRC risk . Gene expression analyses indicated that a ) SNPs identified in our study impact HIATL1 expression , b ) HIATL1 is involved in signaling pathways related to CRC and expression differs between normal and tumor CR tissue , and c ) HIATL1 expression in colon tissue differs by alcohol consumption . The most significant variant rs9409565 is correlated with 142 variants ( LD r2≥0 . 5 in Phase 3 1000 Genomes European populations ) , which spanned across intronic regions and approximately 50kb downstream and 75kb upstream of HIATL1 . Nine of these variants ( including rs9409550 , rs4744345 , rs9409546 , rs9409778 , and rs639276 , all with interaction P<5×10−8 ) fall within a transcriptionally active region in normal colon , rectal and duodenal mucosa [44] as defined by epigenetic signals . [45] Furthermore , these variants fall in a region of enriched enhancer signal; although we note that currently available ChIP-seq data are not able to identify a putative transcription factor binding site at any of the tagged SNPs ( S6 Fig ) . In support of our findings that HIATL1 expression is higher in tumor than adjacent normal colorectal tissue , ChIP-seq ( H3k27ac ) enhancer signals suggest that this locus implicates a gained enhancer present in CR tumors that is absent in normal crypt cells ( S6 Fig ) . In summary , multiple data points suggest that the genetic variants we identified to interact with alcohol on CRC risk are located in regulatory regions impacting the expression of HIATL1 and that HIATL1 expression varies by alcohol consumption . HIATL1 is a member of the solute carrier ( SLC ) group of membrane transport , which enables the directed movement of substances ( such as peptides , amino acids , proteins , metals , and neurotransmitters ) into or out of cells and plays an important role in a variety of cellular functions [46 , 47] . Although the detailed function of HIATL1 remains elusive , this gene was found to be expressed in a large range of animal species and it is highly evolutionarily conserved [48] , suggesting an potentially important functional role . Transporter proteins are commonly upregulated in many cancers [49 , 50] and take part in nutrient signaling to the mTOR pathway [51] which is an important signaling pathway in apoptosis and cancer [52–54] . Alcohol may modify the effects of HIATL1 on CRC risk through its influence on the gene expression of HIATL1 . Nonetheless , the precise mechanism ( s ) of the interaction between alcohol and HIATL1 on CRC risk remains unclear and further studies are needed . Our Cocktail method for detecting G×E interactions did not identify the statistical interaction detected by the conventional logistic regression analysis because rs9409565 did not show strong statistical evidence for association with CRC risk in the marginal association analyses ( P = 0 . 54 , OR = 1 . 014 ) or with alcohol consumption ( P = 0 . 22 ) . Accordingly , this SNP was ranked low in step 1 of the Cocktail method , resulting in very stringent alpha-threshold for the interaction term in step 2 . Although the conventional logistic regression analysis tends to be less powerful overall for genome-wide interaction analysis compared with the Cocktail method [14 , 55] , it has greater power to detect an association if the marginal association of the SNP on disease or the correlation of the SNP with environmental factor are weak as it was the case for the observed interaction . In addition , no association between rs9409565 and alcohol consumption excluded the possibility that the observed interaction was due to the dependence between them [56] . We also explored the effect of rs9409565 and alcohol using other potentially more powerful single step approaches and observed a similar interaction effect in the Empirical Bayesian analysis[57] and a weaker interaction effect in the case-only analysis[58] , which may be explained by the non-significant differential effect of alcohol on CRC in individual carrying the CC genotype ( S6 Table ) . To investigate if genome-wide interaction may help identifying variants that would be missed we looked up the marginal association of rs9409565 in the largest GWAS[59] which is about twice as large as our study and showed an OR for rs9409565 of 0 . 975 ( 95%CI 0 . 946–1 . 007 , p-value 0 . 127 ) . Accordingly , the variant by itself showed only weak evidence for association with CRC . This may not be surprising given that it is estimated that the sample sizes required to identify GxE interaction vs . main effects is at least 4x larger[60] . Our study has several strengths , including the large sample size , environmental exposure assessment in well-characterized populations , and standardized harmonization of environmental data across studies . Further , there is no evidence of heterogeneity across studies for our findings , indicating our results are not dominated by one or a few studies and , indeed , represent evidence across all studies . There are also some limitations . Because amassing sufficient study power for genome-wide interaction analysis is a challenge , we combined all studies in the analysis to gain the greatest power[61] instead of dividing studies into discovery and replication sets . Although we do not have a replication set , the consistency of our findings across all studies and the independent evidence from different types of gene expression data and bioinformatics analyses support a novel interaction for CRC risk between alcohol intake and variants in the 9q22 . 32/HIATL1 region . Our analyses focused on current alcohol consumption , rather than lifetime alcohol use , which may cause misclassification of a certain portion of alcohol users . Both differential and non-differential misclassifications of alcohol consumption levels tend to lead to underestimation of interaction parameters ( e . g . leading to non-significant interaction term between SNP and alcohol intake ) [62] , accordingly , we may have missed some true interactions . However , it is unlikely that this led to false positives for the interactions observed . Because , there is no strong evidence that the type of alcohol ( usually defined as wine , beer and hard liquor ) has a differential impact on CRC[63] we have not investigated interaction between genetic variants and type of alcohol . As we preformed genome-wide interaction testing for two environmental risk factors ( smoking and alcohol consumption ) , additional adjustment for multiple comparisons may be needed . However , we note that the observed interaction at 9q22 . 32/HIATL1 would remain borderline significant ( alpha threshold = 5×10−8/2 = 2 . 5×10−8 ) . The small numbers of heavy drinkers , particular in women , impeded the reliable estimation of interaction parameters and limited our power to identify significant interaction between SNP and heavy drinking . We focused gene expression analysis on HIATL1 because rs9409565 is located in an intergenic region between HIATL1 and FBP2 and further there is a recombination hotspot lying between rs9409565 and FPB2 . If we expand gene expression analyses for all genes 500kb upstream or downstream 500kb of rs9409565 in the 35 pairs of colorectal tumor-normal tissue samples ( S2 Text ) we observed no significant result after false discovery rate ( FDR ) correction . The most significant results were for MIRLET7F which has a p value of 0 . 001 for testing differential gene expression across various levels of lifetime alcohol consumption in normal tissues and PTPDC1 which has a p value of 0 . 002 for testing differential gene expression across various levels of alcohol consumption at reference time . Further studies are needed to confirm our findings . Alcohol has a particularly detrimental effect on several cancers , possibly including CRC , in Asian subpopulations with genetic determined alcohol sensitivity[64–66] . However , as we have focused our analysis on European descent populations and did not observe significant differences of the alcohol-CRC association between studies ( phet = 0 . 16–0 . 76 ) we do not expect major underlying differences of the effect of alcohol in our study populations . We did not perform stratification analyses by anatomical sites for our genome-wide GxE interaction analysis because the association of CRC with alcohol consumption ( S7 Table ) and smoking [23] did not vary according to anatomical site within the large bowel . Although we did observe potential interactions for alcohol consumption , we did not observe statistical evidence for genome-wide SNP x smoking interactions . This may be because smoking has a weaker association with CRC compared with alcohol intake [24 , 26 , 67] , so we may have been underpowered even with more than 10 , 000 cases and 10 , 000 controls . We also may not have properly captured the most relevant smoking variables , such as duration of smoking or time since quitting smoking . The association between smoking and CRC risk are strongest for tumors that display certain molecular features such as microsatellite instability ( MSI ) -high and CpG island methylator phenotype ( CIMP ) -positive [68 , 69] . Because of the lack of MSI or CIMP data in several studies , we cannot perform stratification analysis by tumor characteristics for smoking-related analyses . We note that it would be too early to make any recommendation on alcohol intake from our findings even after independent replication given that such recommendation need to be considered in context of the effect of alcohol on all diseases . Furthermore , it will be important to investigate the interactions between alcohol and genetic variants in larger studies to comprehensively evaluate the full impact of genetic variation on the effect of alcohol on colorectal cancer risk . In summary , we identified a tentative novel interaction for CRC risk between alcohol intake and variants at 9q22 . 32/HIATL1 . Further replication and functional studies are required to confirm our findings and understand the biologic implications of the interaction . This , in turn , could provide further insight into CRC etiology and may identify potentially susceptible subpopulations .
The overall project was reviewed and approved by the Fred Hutchinson Cancer Research Center Institutional Review Board ( approval number: 6501 and 3995 ) . Each study was approved by the local IRB [University of Hawaii Human Studies Program ( Colo23 and MEC ) ; University of Utah Institutional Review Board ( DALS ) ; Partners Human Research Committee ( NHS and PHS ) ; Harvard School of Public Health Institutional Review Board ( HPFS ) ; Fred Hutchinson Cancer Research Center Institutional Review Board ( VITAL , overall study ) ; Ethics Committee of the Medical Faculty of the University of Heidelberg ( DKFZ ) ; NCI Special Studies Institutional Review Board ( PLCO ) ] . For each participating study , participants or the next of kin in the case of deceased participants , provided either written informed consent to participate ( Colo23 , DACHS , DALS , MEC , PHS , PLCO , VITAL , WHI ) or they provided implied written consent by the return of the mailed questionnaires ( NHS , HPFS ) . Additional consent to review medical records was obtained through signed written consent . We included 14 study centers from the CCFR and GECCO as described in the S1 Text and S1 and S2 Tables . All colorectal cancer cases were defined as colorectal adenocarcinoma and confirmed by medical records , pathologic reports , or death certificates . We included advanced colorectal adenoma , a well-defined colorectal cancer precursor [70 , 71] , from two studies ( S1 Text ) . Advanced adenoma was defined as an adenoma 1 cm or larger in diameter and/or with tubulovillous , villous , or high-grade dysplasia/carcinoma-in-situ histology . Colorectal adenoma cases were confirmed by medical records , histopathology , or pathologic reports . Controls for adenoma cases had a clean sigmoidoscopic or colonoscopic examination . All participants provided informed consent and studies were approved by their respective Institutional Review Boards . Average sample and SNP call rates , and concordance rates for blinded duplicates have been previously published [3] . In brief , genotyped SNPs were excluded based on call rate ( < 98% ) , lack of Hardy-Weinberg Equilibrium in controls ( HWE , p < 1 x 10−4 ) , and low minor allele frequency ( MAF<0 . 05 ) . We imputed the autosomal SNPs of all studies to the Northern Europeans from Utah ( CEU population ) in HapMap II . SNPs were restricted based on per-study minor allele count > 5 and imputation accuracy ( R2 > 0 . 3 ) . After imputation and quality-control ( QC ) exclusion , approximately 2 . 7M SNPs were used in analysis . All analyses were restricted to individuals of European ancestry , defined as samples clustering with the Utah residents with Northern and Western European ancestry from the CEPH collection population in principal component analysis [72] , including the HapMap II populations as reference . Statistical analyses of all data were conducted centrally at the GECCO coordinating center on individual-level data to ensure a consistent analytical approach . Unless otherwise indicated , we adjusted for age at the reference time , sex ( when appropriate ) , center ( when appropriate ) , and the first three principal components from EIGENSTRAT to account for potential population substructure . The alcohol and smoking variables were coded as described above . Each directly genotyped SNP was coded as 0 , 1 , or 2 copies of the variant allele . For imputed SNPs , we used the expected number of copies of the variant allele ( the “dosage” ) , which has been shown to give unbiased test statistics [74] . Genotypes were treated as continuous variables ( i . e . log-additive effects ) . Each study was analyzed separately using logistic regression models and study-specific results were combined using fixed-effects meta-analysis methods to obtain summary odds ratios ( ORs ) and 95% confidence intervals ( CIs ) across studies . We calculated the heterogeneity p-values using Woolf’s test [75] . Quantile-quantile ( Q-Q ) plots were assessed to determine whether the distribution of the p-values was consistent with the null distribution ( except for the extreme tail ) . Subjects with missing data for SNPs or environmental factors were excluded from the relevant analyses . Considering the potential male-female difference in alcohol metabolism[76 , 77] and the different levels of alcohol consumption between sexes , we conducted the genome-wide interaction analysis for alcohol separately for men and women and used fixed effects meta-analysis to combine their results . All analyses were conducted using the R software ( Version 3 . 0 . 1 ) . Two statistical methods that leverage SNPs and environmental factors interaction ( G×E interaction ) were used to detect potential disease associated loci . First , we used conventional case-control logistic regression analysis including G×E interaction term ( s ) . As the alcohol consumption variable has three categories there are two interaction terms in the statistical models . Based on an increasing number of publications [78–83] providing a detailed discussion on the appropriate genome-wide significance threshold , which all arrive at similar values in the range of 5 x 10-7to 5 x 10−8 for European populations , we decided to use an alpha level of 5 x 10−8 as the genome-wide significance threshold , assuming about 1 million independent tests across the genome ( 0 . 05/1 , 000 , 000 = 5 x 10−8 ) . For significant results we used permutation approach to determine the empirical p-value . We defined the number of permutation needed as 1/p-value ( i . e . , for a p-value of 5 x 10−8 1/5E-08 = 20 , 000 , 000 ) . We permutated the case-control status 1/p-value times and calculated the p values for the interaction from each meta-analyses to calculate the permuted p-value . Second , we used our recently developed Cocktail method . [55] In brief , this method consists of two-steps: a screening step to prioritize SNPs and a testing step for GxE interaction . For the screening step , we ranked and prioritized variants through a genome-wide screen of each of the 2 . 7M SNPs ( referred to as “G” ) by the maximum of the two test statistics from marginal association testing of Gs on disease risk [84] , and correlation testing between G and exposure ( E ) in cases and controls combined . [85] Based on the ranks of these SNPs from screening , we used a weighted hypothesis framework to partition SNPs into ordered groups and assigned each group an alpha-level cut-off , with higher ranked groups from the screening stage having less stringent alpha-level cut-offs for interaction [86 , 87] . The second step of the Cocktail method is the testing step . We used either case-control ( CC ) or case-only ( CO ) logistic regression to calculate a p-value for the interaction . If the G was assigned based on its low marginal association P value in the screening tests , we used CO test; if it was ranked because of a low correlation screening p-value , we used CC tests . We compared the test step p-value to the alpha-level cutoff for each SNP in a given group . We calculated absolute risks for each genotype of the SNP showing significant G×E interaction . Briefly , based upon the Surveillance , Epidemiology , and End Results ( SEER ) age-adjusted colorectal cancer incidence rate ( denoted by “I” ) between 1982–2011 among the White population of 42 . 9 per 100 , 000 men and women per year , we estimated the reference incidence rate of colorectal cancer ( denoted by “I_{reference}” ) using the following formula: I_{reference} = I/ ( P ( AA , non-E ) + OR{Aa , non-E}×P ( Aa , non-E ) + OR{aa , non-E}×P ( aa , non-E ) + OR{AA , E}×P ( AA , E ) + OR{Aa , E}×P ( Aa , E ) ) + OR{aa , E}×P ( aa , E ) ) , where P ( genotype , E ( or non-E ) ) is the prevalence of light-to-moderate drinking ( or non/occasional drinking ) in each corresponding genotype category among controls ( non-cases ) . Based on this reference incidence rate of colorectal cancer ( i . e . , I_{reference} ) , we further calculated absolute colorectal cancer incidence rates within each subgroup defined by genotype of the SNP according to a light-to-moderate drinking or non/occasional drinking by multiplying the I_{reference} with each corresponding OR . Bootstrap methods were used to calculate the 95% CI of absolute risk estimates [88] . We used different types of gene expression data to examine putative expression of genes identified in our genome-wide interaction analysis , and to determine biological plausibility that the variants identified might impact CRC risk . First , we searched the Genotype-Tissue Expression project ( GTEx ) portal ( http://www . broadinstitute . org/gtex/searchGenes ) [34] and the Human Protein Atlas ( http://www . proteinatlas . org ) [35] to establish whether the implicated genes and corresponding proteins are expressed in human colon/rectal tissues . Second , we used several eQTL databases including the Browser at University of Chicago ( http://eqtl . uchicago . edu/Home . html ) , the Genevar ( GENe Expression VARiation ) at the Wellcome Trust Sanger Institute ( http://www . sanger . ac . uk/resources/software/genevar ) [42] , HaploReg ( http://www . broadinstitute . org/mammals/haploreg/haploreg . php ) ( PMID:22064851 ) , and the GTEx Portal Version 4 ( http://gtexportal . org/home/ ) ( PMID: 26484569 ) to investigate whether any of the implicated SNPs may impact the expression of the nearby genes . A cis-eQTL analysis was also performed in TCGA COAD data in 356 Caucasian samples that have demographic and clinical data for 15 , 008 genes ( S1 Text ) . Third , we analyzed expression data for the implicated genes from 35 pairs of colorectal tumor-normal tissue samples included in the ColoCare Cohort ( S2 Text ) as well as expression data from the Cancer Genome Atlas ( TCGA; http://cancergenome . nih . gov ) in 50 pairs of colorectal adenocarcinoma-normal tissue samples . We searched the UCSC Cancer Genomics Browser ( https://genome-cancer . ucsc . edu ) [37–39] to examine whether the implicated genes showed evidence of differential expression in colorectal tumor tissue and normal tissue . Last , we used the publically available data in the Gene Expression Omnibus site ( http://www . ncbi . nlm . nih . gov/geo/ ) [89 , 90] and the gene expression data from normal colon ( n = 33 ) and tumor ( n = 28 ) tissue in the ColoCare Cohort ( S2 Text ) to investigate whether the expression of implicated genes are correlated with alcohol/smoking history . We explored potential functional annotations for the SNPs that showed evidence for interactions with either smoking or alcohol in our genome-wide interaction analyses . As detailed in S1 Text , we queried multiple bioinformatics databases using the UCSC genome browser ( http://genome . ucsc . edu ) , HaploReg ( http://www . broadinstitute . org/mammals/haploreg/haploreg . php ) , and literature review of published enhancer signatures of colon cancer . | Alcohol consumption and smoking are associated with CRC risk . We performed a genome-wide analysis for interaction between genetic variants and alcohol consumption and cigarette smoking to identify potential new genetic regions associated with CRC . About 8 , 000 CRC cases and 8 , 800 controls were included in alcohol-related analysis and over 11 , 000 cases and 11 , 000 controls were involved in smoking-related analysis . We identified interaction between variants at 9q22 . 32/HIATL1 and alcohol consumption in relation to CRC risk ( Pinteraction = 1 . 76×10−8 ) . If replicated our suggested finding of the interaction between genetic variants and alcohol consumption might contribute to understanding colorectal cancer etiology and identifying subpopulations with differential susceptible to the effect of alcohol on CRC risk . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genome-wide",
"association",
"studies",
"medicine",
"and",
"health",
"sciences",
"cancers",
"and",
"neoplasms",
"diet",
"habits",
"oncology",
"nutrition",
"genome",
"analysis",
"smoking",
"habits",
"digestive",
"system",
"genetic",
"epidemiology",
"genomics",
"behavior",
"epidemiology",
"gene",
"expression",
"adenomas",
"alcohol",
"consumption",
"gastrointestinal",
"tract",
"colorectal",
"cancer",
"anatomy",
"genetics",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"colon",
"human",
"genetics"
] | 2016 | Genome-Wide Interaction Analyses between Genetic Variants and Alcohol Consumption and Smoking for Risk of Colorectal Cancer |
Decades ago , the “immortal strand hypothesis” was proposed as a means by which stem cells might limit acquiring mutations that could give rise to cancer , while continuing to proliferate for the life of an organism . Originally based on observations in embryonic cells , and later studied in terms of stem cell self-renewal , this hypothesis has remained largely unaccepted because of few additional reports , the rarity of the cells displaying template strand segregation , and alternative interpretations of experiments involving single labels or different types of labels to follow template strands . Using sequential pulses of halogenated thymidine analogs ( bromodeoxyuridine [BrdU] , chlorodeoxyuridine [CldU] , and iododeoxyuridine [IdU] ) , and analyzing stem cell progeny during induced regeneration in vivo , we observed extraordinarily high frequencies of segregation of older and younger template strands during a period of proliferative expansion of muscle stem cells . Furthermore , template strand co-segregation was strongly associated with asymmetric cell divisions yielding daughters with divergent fates . Daughter cells inheriting the older templates retained the more immature phenotype , whereas daughters inheriting the newer templates acquired a more differentiated phenotype . These data provide compelling evidence of template strand co-segregation based on template age and associated with cell fate determination , suggest that template strand age is monitored during stem cell lineage progression , and raise important caveats for the interpretation of label-retaining cells .
How stem cells maintain genetic and epigenetic constancy throughout repeated divisions is currently unknown . According to the “immortal strand hypothesis” [1] , as the stem cell divides asymmetrically , it selectively retains those sister chromatids containing the older template DNA strands in the daughter destined to be the renewed stem cell , thus passing the younger strands ( with any mutations acquired during replication ) , to the tissue-committed cell . This phenomenon of template strand segregation was originally based on observations in embryonic fibroblasts [2] and supported by evidence from dividing cells in the intestinal epithelium [3] . Little additional evidence in support of this hypothesis was reported until recently when the immortal DNA hypothesis was revisited , and evidence in support of this process was detected in vitro in immortalized mouse tumor cells [4] and neurosphere cultures [5] , and in vivo in intestinal [6] , mammary [7] , and muscle [8] stem cells . However , the in vivo examples of strand segregation have been limited to at most a few percent of the cells . Thus the phenomenon has yet to be broadly accepted and is attributed to a curious , but minor , cell population . In studies ( unpublished data ) of the timing of proliferation and renewal of skeletal muscle stem cells , or “satellite cells , ” we used different halogenated thymidine analogs ( bromodeoxyuridine [BrdU] , chlorodeoxyuridine [CldU] , and iododeoxyuridine [IdU] ) delivered at different times during regeneration to label sequential cell divisions . To our surprise , although proliferating cells incorporated both labels when we sequentially delivered two of the analogs , approximately half of the cells that ultimately returned to quiescence contained only the second label . Theoretically , this could be explained by the ability of the self-renewing cells to selectively retain the sister chromatids with the older , unlabeled template strands , consistent with the immortal strand hypothesis [1] . We thus examined myogenic progenitors during regeneration for direct evidence of segregation of older and newer template strands .
Muscle was injured to induce regeneration , and 2 d later , pulses of CldU were administered followed by pulses of IdU approximately 12 h later . As such , cells were labeled with CldU during one replicative cycle , and with IdU during the subsequent round of DNA replication ( Figure 1 ) . Cells were then isolated , plated singly , and after allowing a short time for individual cells to complete mitosis , the cells were fixed and immunostained for CldU and IdU . The cell pairs were clearly of the myogenic lineage because this procedure yields cell pairs that are nearly all positive for Syndecan-4 and Pax7 , well-established myogenic markers [9] , and the pairs are clearly replicating as demonstrated by expression of Ki67 ( Figure 2A ) . Daughter cell pairs were analyzed for the distribution of the two labels . Nearly all the pairs of cells were labeled with IdU , confirming that they had undergone DNA replication during the more recent IdU pulse . However , strikingly , we observed asymmetric inheritance of CldU , with all of the detected label in only one daughter cell ( Figure 2B ) , indicating that during the final cell division , one daughter cell had excluded those chromatids containing the template DNA that was labeled during the earlier cell division ( see Figure 1 ) . Even more striking was the fact that this was not a rare event whatsoever; this occurred in nearly half of the pairs ( Figure 2C ) . We also examined the inheritance of labeled DNA in an ex vivo system in which satellite cells activate and proliferate while still associated with individual muscle fibers in culture [10 , 11] . We again observed a very high frequency of co-segregation of labeled chromatids ( Figure 2C ) . By contrast , when the same labeling procedure was applied to proliferating myoblasts , asymmetric segregation of label was seen in only 5% of pairs ( Figure 2C ) . Although markedly less frequent than in satellite cells activated in vivo or ex vivo , this still reflects a mechanism that is maintained in myogenic cells throughout replicative expansion , perhaps by the presence of the few stem-like cells that are propagated in myoblast cultures [12 , 13] . This suggests that the mechanisms underlying template strand segregation may be most active in cells maintained in the stem cell niche . As a further confirmation that activated satellite cells asymmetrically segregate chromatids based on template age , we used a single-label protocol . Muscle was injured and BrdU was injected early during regeneration , comparable to the timing of the CldU pulse in the previous studies . The following day , after cells that had incorporated BrdU would have divided , giving rise to two BrdU-labeled daughters , cells were isolated , plated singly , and treated with either cytochalasin D or nocodazole for several hours to arrest cytokinesis . The cells were fixed and immunostained for BrdU to test for asymmetric segregation of the BrdU label . Of the pairs showing any BrdU label , again about half clearly showed only one daughter cell inheriting the BrdU ( Figure 2D ) . According to the immortal strand hypothesis , the daughter cell inheriting the older template DNA is the renewing stem cell , whereas the other daughter acquires a more differentiated phenotype . However , our experimental protocol was not designed to test specifically for self-renewal , but rather focused on the proliferative expansion and myogenic lineage progression of the stem cell progeny . During myogenic lineage progression , satellite cells differentiate into fusion-competent , Desmin-expressing myoblasts [14 , 15] . We examined cell pairs exhibiting asymmetries in inheritance of DNA templates for the expression of Desmin to test if one daughter cell of each pair was more differentiated than the other and if specific templates segregated with specific cell fates . Pairs that showed asymmetric BrdU staining and any evidence of Desmin expression were further characterized and quantified . Strikingly , the great majority of pairs ( 79% ) showed Desmin expression only in the daughter inheriting BrdU-labeled templates ( Figure 3A and 3B ) , indicating a direct correlation between the inheritance of the younger template and the acquisition of a more differentiated fate , consistent with the underlying assumptions of the immortal strand hypothesis . Much smaller percentages of such pairs showed either asymmetric Desmin expression , but with the Desmin expressed in the BrdU-negative cell ( 18% ) , or symmetric Desmin expression ( Figure 3B ) . This suggests that template strand co-segregation may not be limited to stem cell self-renewal , but may in fact occur more generally during stem cell expansion when asymmetric cell divisions or divergent daughter cell fates are determined . Of the pairs in which BrdU-labeled templates were symmetrically inherited ( about 50% of total cell pairs ) , nearly all were also symmetric for Desmin expression , either both positive ( 59% ) , reflecting symmetric divisions of myoblasts , or both negative ( 31% ) , reflecting symmetric divisions of early progenitors ( Figure S1 ) . Only a very small minority of pairs ( 9% ) with symmetric BrdU showed asymmetric Desmin expression ( Figure S1 ) . Several studies have identified Sca-1 as a marker of undifferentiated progenitors derived from skeletal muscle , demonstrating that satellite cells can give rise to progeny that express Sca-1 at least transiently [16–18] . Because of the treatments needed to detect both BrdU and Desmin immunohistochemically , we were not able to detect Sca-1 in populations that were also stained for both Desmin and BrdU . We could , however , detect clear Sca-1 immunostaining under milder conditions and compare its expression with Desmin in cell pairs . Given the strong correlation between asymmetric Desmin and asymmetric BrdU staining ( Figure 3A and 3B ) , we used asymmetric Desmin as a surrogate marker of asymmetric inheritance of labeled template strands . Among pairs with asymmetric Desmin , the vast majority ( 84% ) also showed asymmetric Sca-1 expression , with Desmin and Sca-1 being mutually exclusive ( Figure 4A and 4B ) . This finding is consistent with the immortal strand hypothesis prediction that the cell inheriting the older template ( in this case , the Desmin− cell ) is the more undifferentiated cell as reflected by the expression of Sca-1 . Only very rarely were pairs detected in which Desmin was expressed asymmetrically and Sca-1 was expressed ( whether asymmetrically or symmetrically ) in the Desmin+ cell ( Figure 4B ) . Virtually all pairs expressing Desmin symmetrically did not express Sca-1 ( Figure S2 ) , consistent with high Desmin expression specifying a more differentiated myoblast and Sca-1 expression reflecting a more immature progenitor . Using these paired cell assays , our data are thus supportive of the immortal strand hypothesis and suggest that template strand segregation is occurring in a large percentage of satellite cell progeny coincident with cell fate decisions . The more immature , Sca-1+ cells undergo asymmetric divisions in which the oldest ( unlabeled , in our studies ) templates segregate to the daughter that retains the less differentiated phenotype as reflected by Sca-1 expression . The other daughter , by contract , acquires the newer templates ( labeled , in our studies , with CldU in the double-label experiments [Figure 2] or BrdU in the single-label experiments [Figure3] ) and adopts a more differentiated phenotype as reflected by Desmin expression . These results would predict that the percentage of Sca-1+ cells that are also CldU+ ( using the double-label protocol ) would decrease through a round of cell division , whereas the percentage of Sca-1− cells that are also CldU+ would increase . To test this directly , we performed experiments as in Figure 2 , but at the time of isolation , half of the cells were fixed immediately as a “before division” snapshot of the population . The other half was cultured for an additional 12 h before harvest and fixation as the “after division” population . Cells were then immunostained for Sca-1 , CldU ( the presumed younger template ) , and IdU ( incorporated into the complimentary DNA strands of divided cells ) , and analyzed by fluorescence-activated cell sorting ( FACS ) . The proportion of cells expressing Sca-1 and labeled with CldU decreased substantially during this time , whereas the Sca-1− , CldU+ proportion increased by a corresponding percentage ( Figure 5 ) . This is consistent with parental Sca-1+ cells segregating older ( unlabeled ) and younger ( CldU-labeled ) templates into two daughters that acquire different fates . These results are also consistent with the proportions of cells asymmetrically inheriting template strands and expressing Sca-1 that we observed in the paired cell assays above .
We propose a model of muscle stem cell proliferation in which muscle progenitor cells divide asymmetrically to generate both myoblasts and immature , undifferentiated cells ( some of which are likely destined to return to quiescence as replacement satellite cells in vivo ) , and symmetrically in order to expand either the pool of progenitors or fusion-competent myoblasts necessary to promote effective muscle repair . The finding of asymmetric inheritance of template strands in the case of the asymmetric divisions and the association of the older templates and the more undifferentiated phenotype is compelling evidence in support of the immortal strand hypothesis , but extends the association of template strand co-segregation to a much broader range of stem cell lineage decisions than just self-renewal . Mechanistically , our data suggest that there must be an ongoing monitoring of template strand age and a process to segregate those strands according to age in a sequential manner , not merely the existence of one immortal strand . The extraordinarily high frequency of muscle progenitor cells exhibiting template strand segregation during muscle regeneration , as opposed to the low frequencies observed in other in vivo systems [6 , 7] , promises to make this system valuable to study mechanisms of asymmetric inheritance of DNA template strands . The high frequency observed in our in vivo studies may relate to the fact that we analyzed cells for asymmetric inheritance of template strands during the process of tissue repair , whereas other in vivo studies have sought evidence of this process during normal homeostatic turnover of tissues [6 , 7] . In addition to providing strong support for the immortal strand hypothesis and expanding the scope of that hypothesis , the findings presented here have additional important implications . First , the assessment of the proliferation kinetics of stem and progenitor cells has been carried out in many tissues by analyzing the dilution of label incorporated into DNA [19 , 20] . The ability of stem or progenitor cells to segregate all label to only one daughter would clearly confound the interpretation of all studies that have heretofore assumed equivalent distribution of label to daughter cells and a simple geometric relationship between label dilution and replicative history . Second , our data require a careful analysis of the use of “label retention” to identify stem cells in tissues , based on the assumption that label retention is equated with very long cell cycle times or quiescence [21–24] . Rather , our data suggest an alternative process by which a cell , even a rapidly dividing cell , could take up label and generate label-retaining progeny . If labeled chromatids continue to co-segregate through repeated rounds of DNA replication and cell division ( see Figure 1 ) , then label-retaining cells can , theoretically , be maintained indefinitely . Accordingly , such a cell would have an indeterminate replicative history since the time the label was administered . The other implication of this caveat is that a label-retaining cell could represent any stage along the lineage from the most undifferentiated stem cell to the most differentiated progeny . In our studies , the label-retaining cell was , in fact , the more committed of the two daughters , and the label-excluding cell was the more undifferentiated of the two . Clearly , the mechanisms that result in label-retaining cells in any tissue may be more complex than simply long cell cycle times , and the relationship between label retention and stage of differentiation may likewise vary from tissue to tissue and under different biological contexts .
Mouse antibody clone B44 recognizing IdU ( and also BrdU ) was obtained from BD Biosciences ( San Diego , California , United States ) ; rat antibody clone BU1/75 ( ICR1 ) recognizing CldU ( and also BrdU ) and rat anti-Sca-1 were from Novus Biologicals ( Littleton , Colorado , United States ) . Mouse and rabbit anti-Desmin antibodies were purchased from Sigma ( St . Louis , Missouri , United States ) and used at 1:200 . Mouse hybridoma supernatant anti-Pax7 was from the Developmental Studies Hybridoma Bank ( http://www . uiowa . edu/~dshbwww/ ) and used at a dilution of 1:5 . Chicken IgY anti-Syndecan-4 was a generous gift from Dr . Brad Olwin ( University of Colorado ) and was used at 1:3 , 000 . Rat anti Ki67 was from DakoCytomation ( Glostrup , Denmark ) and was used at 1:50 . Isotype-matched antibodies were used as controls . Antibody staining was performed as previously described [15] . Unless otherwise indicated , primary antibodies were used at 0 . 5–1 μg/ml . Higher concentrations resulted in detectable cross-reactivity for the antibodies against the halogenated thymidine analogs . Secondary antibodies were Alexa 488- or 546-coupled anti-rat , anti-rabbit , anti-chicken , or anti-mouse antibodies ( Invitrogen/Molecular Probes , Carlsbad , California , United States ) used at 1:2 , 000 for immunofluorescence microscopy . For detection of labeled DNA , cells were fixed in 70% ethanol , washed in PBS , denatured in 2 . 5 M HCl for 30 min , and permeabilized in 0 . 25% Triton-X-100 for 5 min before incubation with primary antibodies overnight in PBS/5% fetal bovine serum . For Pax7 detection , cells were fixed in paraformaldehyde and incubated as above with Triton . For Sca-1 and Syndecan-4 labeling , cells were fixed in 4% paraformaldehyde , washed , and incubated with primary antibody ( in PBS/5% fetal bovine serum for Sca-1; and in 10% BlockHen [Aves Labs , Tigard , Oregon , United States] for Syndecan-4 ) overnight without any detergent permeabilization [17] . Muscle injury was induced by the injection of 1–2 μl of cardiotoxin I ( 100 μg/ml; Sigma ) into 24 sites in muscles of the limb . This produces a diffuse necrotic injury and results in the activation of satellite cells throughout the muscles . BrdU , IdU , and CldU were purchased from Sigma and used at a dose of 30 mg/kg ( subcutaneously ) . For CldU/IdU double-labeling experiments , two doses of CldU were administered 4 h apart , with the first dose administered 48 h after the muscle injury . Approximately 8 h after the second dose of CldU , IdU was administered also by two sequential injections , the second one 4 h after the first . For in vivo BrdU-labeling experiments , two doses were administered 4 h apart , with the first dose administered 48 h after the muscle injury . For in vitro experiments , thymidine analogs were used at a final concentration of 5 μM in the media . For CldU/IdU experiments , either in vivo or in vitro , similar results were observed when the two labels were reversed . As previously described [25] , muscle was dissected , digested in 0 . 25 U/ml collagenase type II ( Sigma ) in HEPES buffered media , and dissociated by trituration into fiber fragments . Fiber-associated cells were liberated either by further digestion in 0 . 5 U/ml dispase and 80 U/ml collagenase in media and then filtration and subsequent washing of cells in PBS , or by trituration in media through a 20-gauge needle [26] . Both methods gave similar results . Cells obtained by this methods are more than 95% positive for the myogenic cell markers CD34 and M-cadherin and less than 2% positive for the endothelial cell marker PECAM [18 , 25] . Cells labeled in vivo and prepared as above were plated at a very low density ( ~10 cells/mm2 ) onto 4- or 8-well chamber slides coated with ECM gel ( Sigma ) diluted to 1:100 . Satellite cells activated ex vivo in bulk myofiber explant cultures were labeled with a pulse for 8 h on day 2 after explantation , maintained in growth medium for an additional 12 h , and then liberated from the fibers and plated singly , as above , early on day 3 . Established myoblast cultures ( passage 20–30 after isolation as bulk cultures [25] ) were labeled , plated singly , and analyzed identically . Direct microscopic examination revealed that sparsely plated cells adhered within about 1 h and that negligible cell migration occurred during the subsequent period before analysis of cell pairs . After cells were attached , cytochalasin D ( 2 μM final concentration; Sigma ) or nocodazole ( 1 μM final concentration; Sigma ) was added to block cytokinesis . Cells were fixed 2–4 h after plating , immunostained , and scored as a “divided pair” if they were within one cell diameter of each other , and more than 50 cell diameters away from other cells in the 20× field of view . Between 100 and 200 cell pairs were scored per experiment , and the number of replicate experiments is described in the figure legends . For co-staining of Sca-1 , CldU , and IdU for FACS analysis , Sca-1 was detected as above using Alexa 647 as the secondary antibody . The samples were then re-fixed , permeablized with 0 . 25% Triton-X-100 , digested with DNAse1 in F-10 medium [8] , and immunostained for CldU and IdU as above , using Alexa 488 or R-phycoerythrin anti-rat secondary antibodies , and PE- or FITC-conjugated mouse anti-BrdU ( IdU ) clone B44 . Isotype-matched antibodies were used as negative controls and for gating . FACS acquisition was performed on a FacsCaliber model ( BD Biosciences ) , and analysis was performed using WinMDI 2 . 8 software ( Joseph Trotter , http://facs . scripps . edu ) . | For each chromosome , the complementary DNA strands consist of a “younger” strand synthesized during the most recent round of DNA replication and an “older” strand synthesized during a previous cell division . When the strands separate to serve as templates for DNA synthesis during a subsequent round of replication , the two sister chromatids formed thus differ in terms of the template strand age . The “immortal strand hypothesis” predicts that a stem cell is capable of distinguishing between chromatids based on template age: when it divides , the self-renewing daughter will inherit the chromatids with the older templates , whereas the daughter destined to differentiate will inherit those with the newer templates . However , in vivo evidence in support of this hypothesis has been sparse . By labeling newly synthesized DNA in sequential divisions of stem/progenitors during muscle regeneration , we observed that almost half of the dividing cells sorted their chromatids based on template age . The more stem-like daughter inherited chromatids with older templates , and the more differentiated daughter inherited chromatids with younger templates . We propose that this phenomenon is a characteristic of asymmetrically dividing stem cells and their progeny . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"in",
"vitro",
"cell",
"biology"
] | 2007 | High Incidence of Non-Random Template Strand Segregation and Asymmetric Fate Determination In Dividing Stem Cells and their Progeny |
Assembly of the ribosome from its protein and RNA constituents has been studied extensively over the past 50 years , and experimental evidence suggests that prokaryotic ribosomal proteins undergo conformational changes during assembly . However , to date , no studies have attempted to elucidate these conformational changes . The present work utilizes computational methods to analyze protein dynamics and to investigate the linkage between dynamics and binding of these proteins during the assembly of the ribosome . Ribosomal proteins are known to be positively charged and we find the percentage of positive residues in r-proteins to be about twice that of the average protein: Lys+Arg is 18 . 7% for E . coli and 21 . 2% for T . thermophilus . Also , positive residues constitute a large proportion of RNA contacting residues: 39% for E . coli and 46% for T . thermophilus . This affirms the known importance of charge-charge interactions in the assembly of the ribosome . We studied the dynamics of three primary proteins from E . coli and T . thermophilus 30S subunits that bind early in the assembly ( S15 , S17 , and S20 ) with atomic molecular dynamic simulations , followed by a study of all r-proteins using elastic network models . Molecular dynamics simulations show that solvent-exposed proteins ( S15 and S17 ) tend to adopt more stable solution conformations than an RNA-embedded protein ( S20 ) . We also find protein residues that contact the 16S rRNA are generally more mobile in comparison with the other residues . This is because there is a larger proportion of contacting residues located in flexible loop regions . By the use of elastic network models , which are computationally more efficient , we show that this trend holds for most of the 30S r-proteins .
Ribosomes are the macromolecular machines that synthesize proteins in all living organisms . They are composed of ribosomal RNA ( rRNA ) and ribosomal proteins ( r-proteins ) that self-assemble into functional units . The efficient and accurate self-assembly of the active ribosome in vivo is essential for cell growth because new ribosomes and proteins must be produced in order for cells to grow . It is estimated that approximately 60% of all cellular transcriptional activities have been attributed to the synthesis of rRNA in a rapidly growing cell [1] and 40% of the total energy of an E . Coli cell is directed towards the synthesis of proteins [2] . It is therefore not surprising that ribosome biogenesis in cells is intricately regulated . Elucidating this complex regulation network has become the focus of a rapidly developing field . The assembly of the ribosome requires the orchestration of highly coordinated events that involve both rRNA folding and r-protein binding . While many cofactors have been identified that participate in assembly in vivo , active functional units can be assembled in vitro in the absence of these cofactors [3] . The small 30S subunit of the bacterial ribosome ( see Figure 1 ) , which is composed of 16S rRNA and 21 r-proteins , has been more extensively studied than other structural assemblages and is a good system to analyze in order to determine what is important for the ribonucleic particle ( RNP ) assembly . In particular , the 30S subunit was the first to be reconstituted from purified components by the Nomura group in the late 1960's [4] . The reconstituted 30S active particles showed nearly the same activities in all performed biochemical assays . This ability to reconstitute active particles in vitro allows for in-depth exploration of the roles of the individual components in ribosome assembly and their functions by the combinatorial addition and omission of individual components [3] , [5]–[6] . These experiments revealed that the 30S subunit assembles in a sequential and ordered process [3] . The Nomura group also provided a detailed assembly map describing the sequential and interdependent binding of all r-proteins [7] . The map also classified the proteins as primary , secondary , and tertiary binders , depending on their ability to bind to 16S rRNA . The primary proteins bind to bare rRNA , secondary proteins can bind to 16S rRNA after at least one primary protein has already bound , and tertiary proteins require at least one primary and one secondary protein [6] . The Nomura assembly map reflects the equilibrium thermodynamics of r-protein binding with 16S rRNA to intermediates . Using chemical probing methods , these binding kinetics were more recently studied by Powers et al . [8] Based on their experimental results , the r-proteins were divided into early , mid , mid-late and late binders . The kinetics data were partially in agreement with thermodynamic data in that the tertiary binding proteins were consistently found to be late binders . The availability of atomic structures of the 30S subunit [9]–[10] provided tremendous new opportunities to understand the assembly mechanism . Most of the knowledge gained in earlier experimental studies was found to be consistent with the determined structures . In the meantime , significant progress was made with experimental methods to probe the ribosome assembly mechanism . Time-resolved X-ray-dependent hydroxyl radical footprinting [11]–[12] provides resolution on the order of milliseconds , much shorter than other chemical probing methods [8] . Directed hydroxyl radical probing [13]–[15] allows for the detection of specific interaction sites between proteins and RNA . The Williamson group used PC/QMS ( pulse-chase followed with quantitative mass spectrometry ) to measure the kinetics of individual protein binding during the assembly of the full ribosomal complex [16] . New experimental data suggest that ribosome assembly proceeds via multiple parallel pathways [16]–[17] rather than a single pathway involving the formation of a single rate-determining “reaction intermediate” RNP [18] . Current understanding of the ribosome assembly process suggests it is similar to protein folding in that it can proceed via multiple pathways across a rugged energy landscape . Many computational studies have shed light on some important aspects of ribosome structure and function . Molecular dynamics simulations have been performed to analyze ribosome interactions with and the accommodation of transfer RNA ( tRNA ) during translation [19]–[22] , as well as to characterize the interactions between cognate tRNA codons and their messenger RNA ( mRNA ) anticodons [23]–[24] . Other simulations and calculations used structures from various stages of translation to study the behavior of incoming mRNA transcripts [25] and nascent polypeptides in the ribosome's exit tunnel [26]–[27] . Interactions between ribosomes and members of a class of antibiotics called aminoglycosides have been elucidated via computational techniques [28]–[31] and have shed light on important interactions between these small molecules and the decoding center of the ribosome . Investigations of the interactions between the ribosome and important non-ribosomal proteins , such as the elongation factor EFTu , have been performed using MD [32] and quantum level calculations [33] . Other quantum calculations have been used to address the function of ribosome catalysis , such as the mechanism of and possible transition states in peptide bond synthesis [34]–[35] . These investigations have enriched the current understanding of ribosomal function and additional computational analyses on the dynamical structure of the ribosome and its components can further elucidate the mechanisms by which the ribosomal machinery assembles and operates . Despite significant progress in recent years , the understanding of ribosome assembly remains limited . One major obstacle in this field is elucidating the mechanisms of coordinated RNA folding , protein binding , and the associated conformational changes of RNA and r-proteins [36] . Although earlier studies suggested [37] that r-proteins adopt the same structures in solution as in the assembled ribosome , more recent studies suggest [36] that there are conformational changes in the r-proteins and rRNA upon forming the complexes . Predicting RNA structure is also one of the most challenging topics in structural biology because a single stranded RNA can adopt a variety of secondary and tertiary structures . The 16S rRNA molecule in a ribosome is divided into four domains: the 5′ domain , the central domain , the 3′ major domain and the 3′ minor , each with a well-defined structure ( see Figure 1 ) . Magnesium ions are thought to stabilize the secondary structure of RNA and many r-proteins are thought to stabilize the tertiary structures . Many of the r-proteins interact with and bind to only one domain , but a few associate with more than one , such as S20 which interacts with both the 5′ and the 3′ minor domains . The Harvey group [38] analyzed the atomic contacts of r-proteins with RNA in the 30S subunit structure and reported the interesting observation that most of the late binding r-proteins were found to bind at the 3′ end of 16S RNA . This observation was consistent with the earlier understanding that 16S RNA folds with 5′ to 3′ polarity [6] , [14] . The Harvey group further used coarse-grained representations of RNP structures to examine the potential fluctuations of binding sites when proteins were removed or added . Their study shows that the binding sites of primary proteins are formed first and , once associated , these proteins help organize the late binding sites . Trylska et al . [39] calculated the binding energy of individual r-proteins with the 16S RNA by solving the Poisson-Boltzmann equation , which accounts for electrostatic interactions . Though the calculated binding energies varied , some late binders were found to have less favorable binding free energies while the early binders were found to be more favorable , an observation consistent with known experimental results . Other studies used various coarse-grained representations to explore the global motions of the ribosome [25] , [40]–[43] and the assembly of the 30S [44]–[45] . Despite the coarse representations of ribosomal structure , some of the known dependencies of r-protein and rRNA binding were captured in these computational studies . Ribosome assembly remains an active research field . A better understanding of its assembly mechanisms will provide valuable biochemical insight into cellular regulation and will allow for the optimal development of ribosome-targeted drugs . While experimental studies continue to make great progress , computational studies reported so far are still limited . Most of the earlier reported computational studies have used coarse-grained representations of the ribosome . To truly understand the specific binding of r-proteins with 16S RNA , atomistic details need to be considered . Because assembly involves both RNA folding and protein binding , the examination of individual components before and after binding in atomistic detail is necessary . Here we specifically investigate the potential correlation between r-protein dynamics properties and their binding properties . The aim is to answer the following specific questions: what are the key residues that bind to the 16S rRNA ? Are these key residues more flexible than the others ? Do free r-proteins adopt the same conformations as those found in the assembled 30S subunit ? To explore the answers to these questions , we rely on the use of atomistic molecular dynamic simulations of r-proteins as well as other methods developed in our own group .
Ribosomal proteins are known to be positively charged and many of these positively charged amino acids , especially those residues on the long extension tails , were found to interact with RNA [10] , [46]–[47] . We performed a simple calculation of the net charge of ribosomal proteins based on the sequences reported for the 2AVY and 1J5E structures , counting Asp and Glu as −1 , Lys and Arg as +1 , with all other residues treated as neutral . Of course , some of these residues might have some charge because of shifted pKa values due to their location in the tertiary structure , but we will ignore these minor effects at present . Table 1 presents the net charge of r-proteins for the two species . The two r-proteins that are not positively charged could be explained by their special positions in the assembly map: S2 is the last protein to assemble [7] and S6 is known to form a dimer with S18 [48]–[49] , which is positively charged , before associating with rRNA . The remaining r-proteins are all positively charged . We also note that the charge on r-proteins from T . thermophilus is on average higher than that for the E . coli proteins , which may relate to the general observation that ribosomal subunits for thermophiles such as T . thermophilus are more stable than those of mesophiles such as E . coli [50] . Moreover , ribosomal proteins are enriched with positively charged amino acids . The typical percent of amino acids for Lys , Arg , Glu and Asp are 5% each for cytosolic proteins [51] . However , in the case of r-proteins , the total percentage of Lys and Arg is approximately 20% ( 18 . 7% for E . coli and 21 . 2% for T . thermophilus ) , while the sum of Glu and Asp percentages remained near 10% . Klein et al had earlier examined the amino acid distributions of r-proteins in the large subunit ( 50S ) and reported a similar bias toward the positively charged amino acids [46] . We have further examined the contacts made between r-proteins and the RNA based on the atomic structures of the 30S subunit from the two species . Here , a contact is defined as having any atoms of a protein residue within 3 . 5 Å of any rRNA nucleotide atoms . Table 2 presents the number of contacts made by each r-protein , along with the number of contacts with positively charged residues . It is clear that a high percentage of contacts between r-proteins and rRNA are made by positively charged residues . The total average percentages of contacts made by positively charged residues are 39% for E . coli and 46% for T . thermophilus , and both are significantly higher than the total percentage of the positively charged amino acids in r-proteins for the two species . These results together affirm the known importance of charge-charge interactions in the ribosome [10] , [46]–[47] . Figure 2 shows structural alignments for the three proteins from the two species . The percentages of sequence identity between the two species are 60% for S15 , ∼40% for S17 , and ∼28% for S20 , but the percentages of conserved residue class are considerably higher: 75% for S15 , ∼58% for S17 , and ∼47% for S20 . Thus , the structures for the three ribosomal proteins are well conserved , with RMSD values of 1 . 1 Å for S15 , 1 . 4 Å for S17 , and 2 . 1 Å for S20 . In the cases of S17 and S20 from T . thermophilus , there are extra C-terminal regions , as shown in Figures 2b and 2c . Residues that contact rRNA exhibit higher than average sequence conservation . For S15 , the percent of conserved contact residues is about 54% ( 52% for E . coli and 56% for T . thermophilus ) , which is just under the overall sequence conservation . For S20 , the percentage of conserved contact residues is 38% for E . coli and 35% for T . thermophilus , both of which are considerably higher than the overall sequence conservation . For S17 , the percentage of conserved E . coli contacting residues ( 52% ) is higher than the overall sequence conservation , whereas that for T . thermophilus contacting residues ( 31% ) is less . The conserved contact residues percentages for S17 and S20 from T . thermophilus are lower than those for E . coli because T . thermophilus has extra C-terminal regions that make several additional non-conserved contacts . ( Supplementary Tables S1 , S2 , S3 present the contact residues for S15 , S17 and S20 for the two species , with conserved residue identities in red and conserved side chain types , largely Lys/Arg substitutions , colored green . ) Further analysis of the identities of these contact residues reveals that , aside from the positively charged residues , His , Thr , Ser , and Gln are also common , all of which are polar and can form hydrogen bonds with rRNA . For example , of the twenty-seven E . coli S15 contacts , five are basic ( Lys48 , Arg54 , Arg64 , Lys65 , and Lys73 ) , five are histidines ( His38 , His42 , His46 , His50 , and His51 ) , ten are polar ( Ser2 , Thr5 , Thr8 , Thr22 , Ser24 , Gln28 , Gln35 , Ser52 , Ser61 , and Gln62 ) , and one is aromatic and polar ( Tyr69 ) . The remaining six contacts are acidic ( Asp21 and Asp49 ) or nonpolar ( Gly23 , Leu31 , Leu39 , and Gly55 ) . Therefore , most contacts between the r-proteins and the rRNA are either charged interactions , or hydrogen bonds , with few aromatic stacking or nonpolar interactions . S15 is a primary binding protein which binds in the 3′ major domain of 16S RNA . In the assembled 30S subunit , S15 is solvent-exposed and located on the back of the 30S subunit body . The 16S RNA binding site of S15 is at the three-way junction of helices 20 , 21 , and 22 in the 16S central domain . The primary , secondary , and tertiary structures of S15 are highly conserved across species: four bundled α-helices are connected by short loops ( Figure 2a ) . All 16S rRNA contact residues are found on one side of S15 , located on helices 1 , 2 and 3 and the loops connecting the three helices , but helix 4 does not have any contacts with rRNA . In previous structural studies , X-ray [52]–[54] and NMR [55]–[56] derived structures were reported and the only significantly different conformation reported was in the crystal structure [52] where helix 1 was rotated 90° away from the remaining bundled helices . Additional studies have been published about the role of S15 in ribosome assembly and antibiotic responses with mutagenesis studies [57] and MD simulations , studying the effects of Mg2+ ions on the protein alone and with its rRNA binding site [56] . It has been suggested that this protein acts as a bridge between the large and small subunits in the fully assembled ribosome [58] . Root-mean-square deviations ( RMSD ) were calculated from the molecular dynamics simulations of the S15 protein and are presented in Figure 3a . The S15 from the two species exhibit relatively low RMSD values during MD simulations , with values remaining below 5 Å . Figure 4 presents the root-mean-square fluctuation ( RMSF ) values calculated over the period of time from 10 ns until the end of the simulation . Contact residues are shown as solid symbols in the plot . High RMSF values were observed for the loop connecting helices 2 and 3 , and several conserved contact residues are located in this loop . The contact residues found on helices 2 and 3 have very low RMSF values , whereas helix 1 and the loop connecting helices 1 and 2 have a few contact residues with moderate RMSF values . Helix 4 , which has no contacts with 16S RNA retains its helical structure during the MD simulation and has moderate RMSF values . Representative backbone structures for E . coli and T . thermophilus S15 are depicted in Figure 5 . The proteins retain their secondary and tertiary structures during the MD simulations and only small conformational changes are observed for either S15 protein . This indicates that the S15 protein from both organisms is a relatively stable protein in solution and that the conformations observed during the simulations are similar to that of the attached protein in the assembled ribosome . Table 3 compares the average RMSF for contact residues with respect to average RMSF for all residues . The average RMSF value for all E . coli S15 residues is 2 . 11 Å and for all contact residues is 2 . 24 Å . For T . thermophilus S15 , all residues average RMSF is 1 . 84 Å and all contacts is 2 . 37 Å . These differences are small , but statistical analysis shows that S15 contact residues are positively enriched with mobile residues , as indicated by enrichment factors greater than 1 for both species ( Table 3; EF = 1 . 08 and p-value = 0 . 217 for E . coli; EF = 1 . 46 and p-value = 0 . 008 for T . thermophilus , see Methodology for explanation of enrichment factors and the p-value ) . The P-values for these enrichment factors signify that the mobility enrichment of T . thermophilus contact residues is significant while it may not for E . coli . In the 30S subunit , S17 is also solvent exposed and is located near S15 in the 5′ domain of the 16S rRNA . To date , no X-ray crystal structures have been determined for S17 alone , but a low resolution NMR solution structure has been presented for Bacillus stearothermophilus S17 [59] . The S17 structure found in the E . coli 30S subunit is comprised of a small β-barrel and an extended ß-hairpin loop ( Figure 2b ) . The contact residues are located on one end of the β-barrel and in the extended ß-hairpin loop . The S17 from T . thermophilus has an extra C-terminal α-helix which makes additional contacts with the 16S rRNA ( Figure 2b ) . Thus , E . coli contact residues exhibit somewhat higher conservation than the overall sequence does , whereas T . thermophilus contact residues are slightly less conserved than the sequence of the full-length proteins . In the E . coli 30S subunit , the S17 ß-hairpin loop is embedded in rRNA and contains five contacts , three of which are found contacting helix 11 of the central domain with two contacting the 5′ domain at helix 21 . The axis of the β-barrel is oriented into the main part of the rRNA , and the end of the barrel nearest the RNA contains the remaining contact points , all of which contact the 5′ domain of 16S rRNA along helices 7 , 9 , and 11 . Because these contacting residues associate with both the 5′ domain and the central domain , E . coli S17 is a plausible anchor between them . The T . thermophilus S17 also contacts these two 16S domains but includes an additional ten protein contacting residues in its C-terminal α-helix and coiled tail . These residues have a larger extent of contact with helix 11 and strengthen the association with the central domain at helices 20 and 27 . Research indicates that the 30S subunit assembly begins at the 16S rRNA 5′ end [8] and , S17 appears to organize the 5′ region [14] , so it is clear that the cooperative conformational changes and rRNA binding of this protein are likely to play an important role in the early stages of ribosome formation . During the MD simulation of E . coli S17 , the β-sheet structures remained stable: the average RMSD for this protein was relatively low ( below 5 Å; lime green plot , Figure 3b ) . Conversely , a much higher RMSD was observed for S17 from T . thermophilus ( olive green plot , Figure 3b ) , although the protein did take on a relatively stable conformation after ∼80 ns of simulation . Further investigation reveals that the extra α-helix in T . thermophilus S17 is responsible for the high RMSD values . The structurally homologous portions of the proteins have comparable RMSD values ( T . thermophilus homolog: dark green plot , Figure 3b ) , both around 4 Å . The backbones of structurally homologous portions both retain their overall shape during the MD simulations . S17 RMSF values ( Figure 6 ) were calculated from the MD simulations starting from the 10 ns point until the end of the trajectory . While the T . thermophilus S17 generally exhibited larger deviations from its starting structure than did the E . coli S17 , when sequentially aligned , the RMSF values for the structurally homologous portions of the proteins correlate well . For E . coli S17 , the loops connecting the ß-strands , the extended ß-hairpin loop , and both termini exhibit comparably high RMSF values , whereas the ß-strands participating in the ß-barrel ( valleys in Figure 6 ) have low RMSF values . The same pattern is true for the homologous portion of the T . thermophilus RMSF plot , and the extra C-terminal region exhibits very large RMSF values . The contact residues in the E . coli S17 are located in the highly mobile ß-hairpin , the moderately mobile Loops 1 and 6 , as well as the least mobile ß-strands of ß-barrel: ß5 , the last residue of ß1 , and the first of ß2 . In T . thermophilus S17 , there are four regions of the protein with high RMSF ( the N-terminus , the ß-hairpin loop , Loop 4 , and the C-terminus ) , all of which contain contact residues . In fact , every residue in Loop 4 is a contact residue , and residues close to each end of the loop also have high RMSF values . The three contact residues in the α-helix have high RMSF and the ten residues in the C-terminal coil have some of the highest RMSF , seven of which are contact residues . The low and moderate contact residues are found in the ß-barrel: ß1 , Loop 1 , ß2 , and ß3 . Representative structures seen throughout the E . coli and T . thermophilus S17 simulations are shown in Figure 7 . The RMSF data and these images indicate that the structurally homologous regions of the S17 protein behave similarly in solution and that the ß structures of both homologs retain their overall shape throughout the simulations , whereas the flexible C-terminal α-helix in T . thermophilus loses its helical structure . These data imply that the ß-barrel confers good stability in solution for the two species . Further analyses of the relative mobility of contact residues shows similar trends as S15 . The average RMSF ( Table 3 ) for all residues in E . coli S17 is 1 . 85 Å and 2 . 28 Å for all contacting residues; for T . thermophilus , the average for all residues is 4 . 68 Å , and 5 . 74 Å for all contacting residues . The differences in these values , while small , indicate that contact residues are , on average , more mobile than all residues for both S17 proteins . Enrichment factors for S17 show positive mobility enrichment for contact residues in both species ( Table 3; EF = 1 . 10 with p = 0 . 199 for E . coli; EF = 1 . 40 with p = 0 . 008 for T . thermophilus ) , with p-values indicating that T . thermophilus enrichment is significant while it may not be for E . coli . In the 30S subunit crystal structures from both species , protein S20 is found deeply embedded in the 16S rRNA . This protein contacts 16S RNA helices 6–9 , 11 , and13 in the 5′ domain and is the only r-protein to contact helix 44 in the 3′ domain . The structure of S20 consists of a unique set of three bundled α-helices , with helix 1 twice as long as the others , the N-terminus most deeply inserted into the subunit , and only a small portion of the three-helix bundle exposed to solvent . While the E . coli and T . thermophilus S20 proteins have a generally conserved tertiary body ( Figure 2c ) , the T . thermophilus S20 crystal structure is missing its first seven residues and has an additional 15 residue C-terminal tail which the E . coli protein does not have . The simulation RMSD values for S20 from both species oscillate wildly ( Figure 3c ) , indicating the proteins conformation vary broadly from their starting conformations ( up to ∼20 Å ) . Multiple length simulations ( at least 200 ns ) show that while S20 RMSD may remain within a range of 5–10 Å for a time , the protein does not adopt a solution-stable conformation . The S20 RMSF plots ( Figure 8 ) have similar trends for both E . coli and T . thermophilus S20 proteins , and aside from the first portion of α1 , the three α-helices are primarily located at valleys in the plots . The highly flexible region of α1 binds to rRNA helices 6 , 7 , and 13 , whereas the nearby , more stable contact residues in α1 contact the tip of rRNA helix 44 , a helix that has no contacts with any other small subunit proteins . The remaining contacts have relatively moderate or low RMSF values . As seen in the other proteins , the loop regions between the stable secondary structures are located at peaks in the RMSF plot , whereas the α-helical regions themselves correspond to the RMSF valleys . Visual inspection of the trajectories suggests that the major contributor to S20 flexibility is helix 1 ( Figure 9 ) , which extends deeply into the rRNA . The N-terminal portion of helix 1 bends and swings wildly during the MD simulations . E . coli helix 1 bends near Arg24 and Thr30 and T . thermophilus near Lys29 . Previous studies [60] have shown that the free S20 protein in solution does not exhibit the high percentage of α-helical regions as seen in the crystallized structure . The conformational variation exhibited by S20 in the work here is consistent with this data , and this flexibility coupled with the deep insertion of the protein into the folds of RNA in the fully-assembled ribosome indicate that S20 is stabilized primarily by its large number of contacts with the RNA . The average RMSF trends ( Table 3 ) for S20 contact residues are generally in agreement with the results presented for S15 and S17 . For E . coli , the average RMSF for all residues is 8 . 82 Å and for all contact residues is 9 . 14 Å . In T . thermophilus , the average value for all residues is 6 . 96 Å and 7 . 62 Å for all contact residues . These data show that the mean RMSF for all contacts is greater than that for the whole structure , consistent with the results for S15 and S17 . Both E . coli and T . thermophilus S20 proteins show positive enrichment of mobility in their contact residues ( Table 3; EF = 1 . 06 with p-value = 0 . 215 for E . coli; EF = 1 . 15 with p-value = 0 . 057 for T . thermophilus ) . However , in this case , the p-values are both greater than 0 . 05 , a typical threshold used for statistical significance test . To rapidly assess the potential connection between contacting residues and their mobilities , we use elastic network modeling which compute RMSF values using only a fraction of the computational resources required for the MD simulations . The elastic network models have been applied previously to the ribosome by us [25] , [40] , [45] , [61] , and in general the dynamics calculated via the Anisotropic Network Model [62]–[63] correlate reasonably well with those from the MD simulations . For example , the correlation coefficient between RMSF values calculated for E . coli S15 is 0 . 57 , for S17 is 0 . 63 , and for S20 is 0 . 81 . ANM and MD predict similar patterns of mobility and stability , with most of the discrepancy at the terminal residues and highly flexible regions ( such as S20 α-helix 1 and S17 ß-hairpin loop ) . In fact , if the first two and last two residues of E . coli S15 are excluded , the correlation factor increases to 0 . 67 . The MD simulations typically predict greater terminal residue mobility ( except for the highly mobile S20 helix 1 ) and the ANM calculations consistently predict higher fluctuation values for extended residues in the middle of the protein . ANM mobility enrichment was calculated for all 19 r-proteins in the two 30S X-ray structures and results are presented in Table 4 . Most r-proteins are significantly enriched for mobile residues at the rRNA contact points at the 0 . 05 level . Contacting residues are not only enriched , but they make up a subset of residues that is near maximal enrichment , for a given structure . Proteins S2 , S6 , S8 , S18 and S19 do not show statistically significant enrichments and are colored red in Table 4 . As mentioned earlier , S2 and S6 differ from the rest of r-proteins in that they do not have a net positive charge . Also S6 and S18 are known to form dimers in solution . Hence calculation of their dynamics as monomers may not reflect their true dynamics in solution . S8 is one of the primary binding r-proteins and S19 is one of the secondary binding r-proteins . At present , we do not know specific properties that may make these two proteins differ from the rest . Although their EF values are greater than one ( rRNA contacts are more mobile ) , their p-values do not reach the level of high statistical significance ( they are not a maximally enriched subset ) . In addition to those r-proteins , S14 , S17 and S20 are not significantly enriched with mobile residues for E . Coli , but are statistically significant enriched for T . Thermophilus . On average , T . thermophilus proteins show a slightly increased enrichment relative to E . coli; with average enrichment factors of 1 . 51 and 1 . 46 , respectively , with medians of 1 . 43 and 1 . 33 . Of the 6 proteins categorized as being early by the Harvey group [38] , two E . coli and five T . thermophilus have mobility enrichments significant at the 0 . 05 level . Of the six primary proteins identified by Nomura [7] , three E . coli and five T . thermophilus are significant at the 0 . 05 level . Proteins involved later in assembly are not differentially significant between the two species . This may imply that thermophiles exhibit increased control over the placement of mobile residues within proteins that bind to rRNA . Several important conclusions can be reached based on the above reported results . First , the positively charged residues on r-proteins must play important roles in binding with 16S rRNA , as noted earlier [10] , [46]–[47] . A significantly higher percentage of contacts between r-proteins and rRNA are formed by these positively charged and hydrogen bonding residues . We also see that r-proteins from a thermophilic species ( T . thermophilus ) have more positively charged residues than a mesophilic species ( E . coli ) , which correlates with the fact that thermophilic ribosomes must maintain stronger ( or a larger number of ) interactions in order to function at considerably higher temperatures . Second , as previously discussed [36] , conformational changes of r-proteins could take place during 16S rRNA binding . Our study clearly shows that α-helix 1 of S20 is unstable in solution by itself and exhibits large conformational changes . In contrast , S15 and S17 adopt stable conformations in solution , which agrees with the earlier suggestion [37] that ribosomal proteins do not undergo structural changes during assembly . We attribute the differences in these behaviors to the extent of solvent exposure the protein experiences within the assembled subunit . In the ribosome , S15 and S17 are primarily solvent exposed so their solution structures would be likely to more closely resemble their bound structures , whereas S20 is deeply embedded in the 16S RNA , and its association with its RNA binding site stabilizes the flexible portion of α-helix 1 . Third , analyses of residue mobilities reveal that RMSF values for contact residues are statistically higher than those for other residues . This means that contacting regions are more enriched with mobile residues than non-contacting regions , which supports previous observations [37] that the flexible regions of ribosomal proteins are usually the locations of RNA contacts . However , this does not mean that all contact residues are located in the flexible loop regions . It is important to point out that there are many contact residues found in α-helices and β-sheets that exhibit low to moderate RMSF values . The trend that contact residues being enriched with mobile residues holds for most of 30S r-proteins , with only a few distinct exceptions like S2 , S6 , S18 . Their exceptions however could be traced to peculiar known facts such as dimerization between S6 and S18 . The increased mobility of contact residues could ensure more efficient binding and even aid in the binding site preparation for later binding proteins by actively associating with their 16S binding partners and helping to fold and maintain the appropriate rRNA tertiary structure . The T . thermophilus exhibited higher enrichment factors than the E . Coli , which may point to a novel adaptation of thermophiles – the increased control over the placement of highly mobile residues .
In the current study , we analyze the crystal structures of the 30S subunits from the Escherichia coli ( PDB [64] ID 2AVY [9] ) and Thermus thermophilus ribosomes ( PDB ID 1J5E [10] ) . Structural and sequence alignments of r-proteins found in the two species were done with Molecular Operating Environment ( MOE ) software ( Chemical Computing Group ) . Contacts between r-proteins and 16S rRNA were analyzed using our own computer program . A contact point was defined as any atom of a protein residue found within a 3 . 5 Å cut-off distance from any 16S nucleotide atom . That amino acid was labeled as a “contact” residue . The total number of “contacts” between one r-protein and the 16S rRNA may exceed the total number of contacting residues identified in the protein because an amino acid may be within cutoff distance of more than one nucleotide , thus counting as more than one contact . The identity and position of these contact residues found in the assembled 30S subunit were recorded and used for further analysis . Molecular dynamics ( MD ) simulations were run using the AMBER 10 software package [65] and the parmbsc0 force field [66] , an optimization of the Amber99 force field for nucleic acids and proteins . The starting conformations of r-proteins for the MD were obtained from the crystal structures of the 30S subunits ( E . coli 2AVY and T . thermophilus 1J5E ) . Counterions were added to neutralize the charge of the protein , and an additional 10 potassium and 10 chloride ions were added to create a low salt concentration . The protein systems were then solvated using a rectangular box of TIP3P water [67] . The systems were subjected to two minimization cycles: 1000 steps with the protein fixed and 5000 steps unrestrained . Afterward , a 100 ps warm-up MD simulation was run at constant volume by increasing temperature from 0 to 300 K , with the protein fixed using a restraint constant of 10 . 0 kcal·mol−1·Å−2 . The MD simulation then switched to the NPT ensemble ( p = 1 . 0 bar ) , using the Langevin thermostat with a collision frequency of 1 . 0 ps−1 , to equilibrate the ions and water density for 2 ns . The restraint force on the protein was then removed and the production run began with the NPT ensemble ( p = 1 . 0 bar ) using a time step of 2 fs . All simulations used the SHAKE algorithm [68]–[69] to constrain covalently bonded hydrogen atoms and the Particle Mesh Ewald ( PME ) method [70] to calculate long-range electrostatic interactions , with a cutoff distance of 10 . 0 Å . Histidines are represented as HIE ( neutral charge: hydrogenated Nε , aromatic Nδ ) . Duplicate MD simulations were performed to verify that the reported dynamic behaviors of each protein are representative in the final MD runs . MD production runs were performed for at least 200 ns , which should be of sufficient length to establish the conformational stabilities of proteins of this size . Using Ptraj to monitor the overall structural changes in reference to the starting structure , the root-mean-square deviation ( RMSD ) for each protein was calculated as a function of production run time . If the plot of the RMSD versus time forms a plateau , the protein likely adopts a solution-stable conformation; however , a widely fluctuating RMSD plot indicates a flexible protein in solution . To quantify the mobility of each residue , root-mean-square fluctuations ( RMSF ) were calculated using the average protein conformation as the reference state . The RMSF values presented in this paper are calculated from 10 ns to the end of each simulation ( approximately 200 ns ) to allow adequate time for the protein to fully adopt its stable solvated conformation , if one was at all achieved . This ensures that the RMSF plot differentiates flexible residues from stationary residues during the time that the protein samples its solution-stable conformations . In both RMSD and RMSF calculations , all atoms were included . The RMSF is related to the experimental B-factors reported by crystallographers , through a simple relationship ( B-factor = ( 8/3 ) π2 ( RMSF ) 2 ) , which could be compared with the experimental measured B-factors reported in the PDB files of the 30S subunits . However , the experimental B-factors for each r-protein found in the 30S subunits were nearly featureless for individual proteins , probably because the reported B factors reflect the mobility of the atoms within the whole assembled subunit and are not representative of the individual r-proteins . Hence , we did not compare the B-factors calculated from MD simulations with the experimental B-factors . Snapshots of each protein at various stages throughout the simulations were visualized using Visual Molecular Dynamics [71] ( VMD ) to identify the flexible and stable regions of the protein . All images were made with VMD , which is developed with NIH support by the Theoretical and Computational Biophysics group at the Beckman Institute , University of Illinois at Urbana-Champaign . Because the Molecular Dynamics simulations require significant resources , we have also chosen to model the dynamics of the complete set of 30S ribosomal proteins with the more computationally efficient elastic network model [72] , using the Anisotropic Network Model in particular [63] , [73] , ANM models permit us to investigate the dynamics of all of the 30S proteins more quickly but with less detail in the observed dynamics than MD , but with greater overall certainty about the large-scale motions of the structures . ANM models are constructed using the crystallographic Cα coordinates of each protein and a cutoff of 13 Å . Due to its coarse-grained design , the ANM is subject to the “tip effect” [74]–[75] in which highly extended points ( Cα ) experience exaggerated motions , which would place disproportionate weight on the most mobile residues . To compensate for this effect , we calculate the RMSF of each residue position in each structure and remove extreme outliers from subsequent analyses . The “tip effect” residues removed in this study are Arg88 and Gly89 from T . thermophilus S15 , and Gly8 , Val9 , Val10 , and Val11 from T . thermophilus S17 . We also use RMSF to make comparisons between 16S rRNA contacting residues and non-contacting or highly conserved residues . The definition of contacting residues and conserved residues is the same in both the ANM calculations and the MD studies . To statistically determine linkages between highly mobile and contacting residues or conserved residues from both ANM calculation and MD simulation , we calculate an enrichment factor for each protein defined as the ratio of the average RMSF for contacting over non-contacting residues . An enrichment factor greater than 1 implies that the contacting residues are more mobile than the non-contacting residues . However , an enrichment factor less than 1 implies the reverse . The statistical significance ( p-value ) of the enrichment factor is calculated based on the permutation test explained as follows . For a protein of N residues , C of which are contacting , we have an observation of the enrichment of RMSF at the contacting residues relative to the non-contacting residues . Let this ratio be O . We then randomly select C residues from the protein and calculate the analogous ratio between this random set and its compliment . Performing the random selection 10 , 000 times , we construct a distribution of enrichment values within random sets of C residues . The significance ( p-value ) of our initial observation , O , is then the proportion of random samples that have an enrichment greater than O . A small p-value ( e . g . , p<0 . 01 ) implies that a random set of C residues is unlikely to have an enrichment factor equal or greater than the observed ratio O . This not only means that the contacting residues are more mobile than the non-contacting residues , but that there are very few subsets of size C exhibiting the same magnitude of mobility . | Ribosomes are complex cellular machines that synthesize new proteins in the cell . The accurate and efficient assembly of ribosomal proteins ( r-proteins ) and ribosomal RNA ( rRNA ) to form a functional ribosome is important for cell growth , metabolic reactions , and other cellular processes . Additionally , some antibacterial drugs are believed to target the bacterial ribosome during its construction . Hence , ribosomal assembly has been an active research topic for many years because understanding the assembly mechanisms can provide insight into protein/RNA recognitions important in many other cellular processes , as well as optimize the development of antibacterial therapeutics . Experimental studies thus far have provided still limited understanding about the assembly process . To further understand the assembly process , we have computationally studied the dynamic properties that r-proteins exhibit during assembly and the relationship between dynamics , physical properties , and binding propensity . We observe significant charged interactions between r-proteins and rRNA . We also detect a strong correlation between contact residues and their dynamic mobilities . Protein residues contacting with rRNA are observed to be more mobile in comparison with other residues . We also relate the location of the r-protein in the fully assembled ribosome to its susceptibility for large conformational changes prior to binding . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Methods"
] | [
"macromolecular",
"complex",
"analysis",
"rna",
"structure",
"biophysic",
"al",
"simulations",
"protein",
"structure",
"biology",
"computational",
"biology",
"macromolecular",
"structure",
"analysis"
] | 2012 | A Computational Investigation on the Connection between Dynamics Properties of Ribosomal Proteins and Ribosome Assembly |
Onchocerciasis , or river blindness , has historically been an important cause of blindness , skin disease and economic disruption in Africa and the Americas . It is caused by the filarial parasite Onchocerca volvulus , which is transmitted by black flies in the genus Simulium . Over the past decade , several international programs have been formed to control , or more recently eliminate onchocerciasis , using mass drug administration ( MDA ) of ivermectin . However , in many areas of Africa ( particularly those which are endemic for the eyeworm , Loa loa , or where vector densities are very high ) ivermectin MDA alone will not be sufficient to achieve elimination . In these situations , additional interventions may be necessary . The Esperanza Window trap ( EWT ) , a simple trap originally developed to replace human landing collections for entomological surveillance of O . volvulus transmission was optimized , resulting in a 17-fold improvement in trap performance . The optimized trap was tested in trials in schools and in agricultural fields to determine if it could reduce vector biting locally . The traps resulted in a 90% reduction in biting in the school setting . In the field setting , results varied . In one location , the traps reduced biting by roughly 50% , while in a separate trial , the traps did not significantly reduce the biting rate . Examination of the two settings suggested that trap placement may be critical to their success . These results suggest that the optimized EWT might be capable of reducing local vector black fly biting in areas commonly frequented by residents . Together with other recently developed methods of community directed vector control , the traps may augment ivermectin MDA , bringing the goal of onchocerciasis elimination within reach in much of Africa .
Onchocerca volvulus , the causative agent of river blindness , remains endemic in most parts of Africa , despite mass drug campaigns with ivermectin spanning more than three decades [1 , 2] . There are multiple causes for this entrenchment , many of which are the result of the difficulty in eliminating a vector-borne disease in general . Some are operational—for example inadequate drug coverage and/or poor timing of mass administration–while others are sociological , i . e . , broad-scale apathy and occasional resistance on the part of individuals and communities to participate in treatment . A complicating factor throughout much of Central Africa is the co-endemicity of O . volvulus with Loa loa . The presence of the latter is particularly important because of severe adverse reactions that may occur in individuals with elevated L . loa microfilaremia following ivermectin treatment [3–5] . This has precluded the use of ivermectin mass drug administration ( MDA ) in many areas and may result in reduced community participation in areas co-endemic for loasis where ivermectin MDA is ongoing [6] . Aside from these operational issues , another fundamental problem is the long-standing hyperendemicity of O . volvulus associated with extremely high annual biting rates by the vectors , members of Simulium damnosum sensu lato species complex . Models show that vector abundance must be considered in situations where annual biting rates are high because treatment with ivermectin may not be able to achieve elimination on its own [7 , 8] . Thus , there is a need for local vector control that is complementary to mass drug administration and is economical , ecologically benign and appropriate for much of rural Africa . The Esperanza Window Trap ( EWT ) was originally developed as a tool to capture vector black flies to supplement or eventually replace human landing collections for verifying the interruption of transmission of O . volvulus in Latin America [9] . Its design was later modified for use in Africa [10] , and shown to be an effective substitute for human landing collections in Uganda [11 , 12] . The EWT , when deployed in households and in schoolrooms in rural Mexico was shown to be capable of significantly reducing the biting rate of the local vector S . ochraceum , suggesting that the EWT might have potential as a mechanism for vector control [13] . In the current investigation , we report the results of a series of studies aimed at optimizing the collection efficiency of the EWT in Uganda and report the results of studies evaluating the ability of the optimized trap to reduce biting of S . damnosum s . l . in two different settings . These studies were carried out in the Madi mid North focus of Uganda , which is the largest focus of onchocerciasis still active in Uganda . Elimination efforts here have lagged behind many of the other foci in Uganda due to political instability in the region and because the vector for O . volvulus in this area is S . damnosum s . l . , which is less amenable to classical vector control measures than S . neavei , the most common vector in Uganda .
All experiments were carried out in the communities of Gonycogo and Laminatoo , located in the Nwoya district of Northern Uganda . Previous studies have indicated that vector density is high in these communities with individuals receiving 52 bites per day in Laminatoo and 100 bites per day in Gonycogo [12] . The location of these villages is shown in Fig 1 . A detailed description of the study villages may be found in our previous publication [12] . The sibling species of S . damnosum s . l . present at these sites is the savanna dwelling cytospecies S . damnosum sensu stricto . The experiments were conducted from May , 2017 through January , 2019 . All traps and landing collections were done on publicly accessible lands . The performance of the different trap designs was evaluated in a pairwise fashion . Three pairs of sites were identified in each village where the traps were set up . The trap sites were at least 50m apart , to ensure that they were independent from one another . However , the traps were kept within 500m of the major black fly breeding site found near each community , to ensure that the traps were sampling from the same population of flies . One trap design was placed at one of each of the pair of sites , while the second design was placed at the second site in each pair . The traps were activated at 7AM by removing a sheet of plastic covering the surface of the trap , and the traps baited with a yeast sugar solution to produce CO2 and a dirty sock , as previously described [12] . The traps were permitted to operate until 6 PM , at which point the flies were removed from the trap surface by placing a small drop of white spirits on each fly to dissolve the glue and removing the fly from the trap with forceps . The flies were then placed in isopropanol , identified morphologically to species and the number of S . damnosum s . s . collected on each trap was recorded . Collections were carried out twice per week at all sites . The position of the trap designs in each pair of sites was switched weekly , to eliminate position effects . The number of trials ranged from 12 to 30 , depending on the trial . The actual number of trials in each experiment is indicated in the figure legend for each pairwise comparison . Because the number of flies collected varied widely from day to day due to weather conditions , the daily collections from each of the designs evaluated in each study were normalized to the total number of flies collected from both trap designs on each day . These studies were conducted in two primary school classes in the village of Gonycogo , Koch in Goma sub-county in the Nwoya district . One of these classes was located in a thatched roofed building open to the outside on one side , while the second class was conducted in the open air under a tree . The school staff were instructed in the purpose of the study . The teachers then briefed the students on the purpose of the study , instructing them to refrain from touching or tampering with the traps . A human landing collector was placed at the periphery of the classes ( so as not to disturb the children ) , and collections were carried out daily in the absence of the traps for a week to establish a baseline biting rate . The human landing collectors operated as a team of two individuals , which alternated collecting every hour throughout the day . Two traps were then set up near each of the classes and collections were continued for an additional week . Collections and traps were operated from 7AM to 5PM on weekdays when classes were in session . Collection numbers were normalized to those obtained from a human landing collection team located roughly 200m from the school , to control for variations in the overall fly population . The studies to evaluate the ability of the EWTs to reduce biting rates in an agricultural setting were carried out in two types of fields in Laminatoo and Gonycogo , one planted with soybeans and one planted with tomatoes . The fields ranged in size from 0 . 5 to 1 . 3 hectares in area . A total of 5 traps were distributed around the periphery of each field and the intensity of vector biting monitored by a human landing collection team placed at the edge of the field . In addition to the collectors located in the test sites , an independent team was set up in each village located approximately 600 meters away from the fields to monitor any changes in the overall fly population . Collections were carried out every other day for a week ( excluding Sunday ) from 7AM through 6PM , at which point the traps were removed . Collections continued to be carried out in the absence of the traps for a week from 7AM through 6PM ( excluding Sunday ) , at which point the traps were replaced . This pattern was continued for a total of five weeks . Fly collection numbers for all studies were recorded using whole day collections as the minimum time unit . To analyze the effect of trap size optimization , the number of flies caught by each trap was evaluated as a proportion of the number of flies caught that day . In the analysis of the unadjusted data , a one-sample Student’s T-test was used in order to determine whether the mean percentage of flies caught by the 1 . 5m traps was significantly different than 50% , i . e . accepting a null hypothesis that the 1 . 5m and 1m traps would be equally effective . In order to adjust for the effect of width , the null hypothesis was taken to be that the proportion of flies collected on the 1 . 5 m trap would be 60% of the total . This was based on the fact that the 1 . 5m trap made up 60% ( 1 . 5m/ ( 1 . 5m + 1 . 0m ) *100 ) of the total trap width tested in the trial . Similarly , in adjusting for area , the null hypothesis was taken that the 1 . 5m trap would collect 69% ( 2 . 25m2/2 . 25m2 + 1m2 ) of the flies . To compare the mean fly capture counts between the 3 x 1m traps and 1 . 5m trap flanked by 2 x 1m traps , a Mann-Whitney test was used due to the non-normal distribution ( Shapiro-Wilk , p = . 029 ) of the data . A Student’s T-test was used in order to determine whether traps with a small black stripe would be more or less attractive when compared to traps consisting of three stripes of equal width . A nonparametric Mann-Whiney test was used for the analysis of stripe shape , due to the non-normal data distribution ( Shapiro-Wilk , p = . 005 ) . Multiple tests were used in the analysis of the data collected in the school study . An ANOVA test with a Dunnett post-hoc was used to determine whether there was any difference in the number of flies caught within the two classrooms and the outside human landing collector . Then , the fly counts within the classrooms were normalized to the number of flies caught by the external human landing collector . A Student’s T-test was then used to determine whether there was a significant difference in the normalized number of flies caught when there was a trap present compared to when there was no trap present . The first step in analysis of the data collected to evaluate the effect of placing the traps in an agricultural setting was to normalize the field collection numbers to those of the external collector . The resulting data were then analyzed using a Mann-Whitney test , due to non-normal distribution ( Shapiro-Wilk , p < . 0001 ) . Data from both villages were analyzed separately . Each analysis was conducted using a significance level of α = 0 . 05 . The experiments described here were reviewed and approved by the Institutional Review Boards of the Uganda Vector Control Division ( approval REF/VCDREC/071 ) and the University of South Florida ( approval CR3_Pro00015108 ) . All individuals participating in the study received Mectizan twice per year as part of the routine mass drug distribution program active in this area administered by the Uganda Ministry of Health .
As a first step in optimizing the performance of the EWT , the effect of the size of the trap was evaluated . To accomplish this , two different trap sizes ( 1m square and 1 . 5m square; Fig 2 , Panel A ) were tested in a pairwise fashion , as described in Materials and Methods . The larger version of the trap was found to collect significantly more flies than did the smaller version ( Fig 2 , Panel B; p<0 . 0001 ) . The advantage conferred by the larger trap was maintained when analysis was adjusted for width ( p< 0 . 005 ) , but this advantage disappeared when analysis was adjusted for trap area ( p = 0 . 65 ) . Thus , the larger traps did not outperform their predicted performance based upon their area alone . This suggested that the increase in the collections seen by using the larger trap would not be any greater than if a number of smaller traps were deployed in an array . To test this hypothesis , a pairwise trial was conducted in which an array of three 1m traps was compared to an array consisting of two 1m traps flanking a single 1 . 5m trap . The collections obtained by the array of three 1m traps were not significantly different than those collected by the array containing the larger trap ( Fig 3; p = 0 . 6 ) . Because the larger traps were unwieldy and difficult to handle under field conditions , all subsequent trials used traps that were 1m square . The initial trials adapting the EWT for use in Africa demonstrated that a striped design was more effective than a solid blue color [10] . Subsequent studies suggested that traps containing one central blue stripe flanked by two black stripes , all of equal width , and traps containing two blue stripes flanking a single black stripe , all of equal width , were equally effective [11] . However , the effect of the width of the stripe had not been explored . To test this hypothesis , a pairwise comparison was carried out comparing the standard design with stripes of equal width to a design with a thinner black stripe ( Fig 4 , Panel A ) . In this case , the traps with the thin stripe collected significantly more flies than did the standard design ( Fig 4 , Panel B; p < 0 . 0001 ) . As a final step in optimizing the EWT platform , the effect of changing the shape of the black stripe was investigated . In this trial , traps with a simple thin black stripe were compared to traps containing a central figure shaped like a human female ( Fig 5 , Panel A ) . It was found that the trap with the human shaped stripe did not collect significantly more flies than did the trap with the simple rectangular thin stripe ( Fig 5 , Panel B; p = 0 . 43 ) . Based upon these findings , all subsequent studies evaluating the EWT as a vector control measure utilized 1m square traps with a thin black stripe . In studies conducted in Mexico , the Latin American version of the EWT when deployed in classrooms reduced biting by S . ochraceum by up to 50% [13] . Therefore , we decided to determine if the optimized version of the EWT described above would be capable of reducing biting in primary school classes in Uganda . In these studies , human landing collections were used to monitor biting rates in two open air classrooms in Gonycogo Primary School in the presence and absence of the traps . The numbers of flies collected in the classrooms were normalized to a human landing collection carried out 200m from the school , as a way of controlling for the day to day variation in the questing fly population . In the absence of the traps , the human landing collections in both classes were consistently higher than those of the external collector ( Fig 6 , Panels A and B ) . In the presence of the traps , this pattern was reversed—the classroom collections were consistently lower than those of the external collector ( Fig 6 , Panels C and D ) . When normalized for the number of flies collected by the external collector , the number of flies collected by the classroom collectors in the presence of the traps was on average 9% of those collected in the absence of the traps ( Fig 6 , Panel E; p < 0 . 0001 ) . The ability of the optimized EWT to reduce biting rates was also evaluated in two agricultural settings in Laminatoo and Gonycogo . In both villages , the traps collected large numbers of flies , with the mean fly collections per trap ranging from 4-fold to 26-fold greater than those collected by the associated human landing collector ( Fig 7 ) . Furthermore , in both the soya and tomato fields in Gonycogo , the number of flies collected in the fields by the human landing collector ( when normalized to those collected by the external collector ) was significantly less when the traps were present than when they were absent ( p = 0 . 006; Fig 8 , Panels A and B ) . In contrast , the biting rates in both fields in Laminatoo were not different when the traps were present when compared to when they were absent ( p = 0 . 25; Fig 8 , Panels C and D ) .
In the initial study carried out in Uganda that compared the EWT to human landing collections for surveillance of O . volvulus transmission , one EWT was found to collect roughly an equivalent number of flies as a human landing collector [12] . In the studies done in the soya and tomato fields of Gonycogo and Laminatoo , each optimized trap collected on average 17 times the number of flies collected by the HLC . The optimization process thus resulted in a dramatic improvement in the performance of the EWT . Despite this , it is likely that additional refinements to the trap might boost the trap performance even further . For example , earlier studies demonstrated that a synthetic bait formulation containing five compounds that are found in the sweat of most individuals previously shown to be attractive to S . damnosum s . l . [14] did not out-perform dirty socks worn by local residents of the study communities [12] . However , it is well known that individuals vary widely in their attractiveness to hematophagous insects [15 , 16] . It is thus possible that the compounds found to be specific to , or in a higher concentration in the sweat of such highly attractive individuals might be used as a way to improve the baits for the EWT . Similarly , the use of alternative methods of CO2 generation [17] , or carbohydrate sources other than sucrose for the generation of yeast produced CO2 [18] might improve trap performance . The performance of the optimized traps varied widely . For example , Trap 3 in the study conducted in the soya field of Gonycogo generally out performed Trap 4; an extreme example of this occurred on May 17 , 2018 when Trap 3 collected 5 , 000 flies and Trap 4 collected just 150 flies . This suggests that placement played a critical role in trap performance , which is consistent with studies carried out in Mexico [19] . Interestingly , the traps were placed in locations that appeared quite similar and all were in lightly shaded areas with good visibility on the edges of the fields . Furthermore , certain locations were not uniformly better than others . For example , trap 3 in the Gonycogo soya field out-performed trap 4 on all days but one , but on June 12 , 2018 , trap 4 collected 8 , 812 flies , while trap 3 collected just 1 , 006 flies . Thus , the performance of the traps varied widely from day to day , and the performance of the traps relative to one another was not consistent . This suggests that a trial and error process may be necessary to identify the locations that optimize trap performance . The optimized EWT appeared to have potential for protecting groups of people from the bites of vector black flies . Before the traps were deployed in the school , the biting rate in both classrooms was significantly greater than that recorded by the external HLC . This in itself was somewhat disconcerting , as it suggests that the normal procedure used to site HLCs might actually underestimate the biting that people gathered in groups suffer . However , when the traps were deployed , the biting rate dropped dramatically in both classrooms . When normalized for the collection obtained by the external HLC , deployment of the traps reduced the biting rate in the classes by 91% . This reduction was greater than that observed in a similar study conducted in Mexico , where deployment of the traps in households and schools reduced biting by 14–51% [13] . This result suggests that the EWT might prove to be an effective means to reduce biting when deployed in key locations where people congregate for such daily activities as school , washing clothes and bathing . In contrast to the dramatic reductions seen in the biting rates in the classrooms , the effects of deploying the EWTs in the fields were less clear . Overall , in Gonycogo a 60% reduction in the biting rate was observed when the traps were present when compared to when they were absent . However , the traps appeared to have no significant effect on the biting rate in the trials conducted in the fields in Laminatoo . One possible explanation for this difference might be in the location of the human landing collectors at each of the fields . In both communities , the human landing collectors were situated in partially shaded locations along the edge of the field . However , when examining satellite images of the fields , it was apparent that the collectors at both of the fields in Gonycogo were located on the western edge of the fields , while the breeding site was located to the east of the fields . Thus , flies coming from the breeding site would have passed the fields and the traps before encountering the collector . In contrast , the collectors in Laminatoo were located on the right edge of the field relative to the breeding site . Here , flies would not have had to travel past the traps before encountering the collector . This suggests that trap position might be very important in achieving a reduction in biting rates , with the traps best positioned so the flies are likely to encounter them before they encounter the humans the traps are protecting . More work will be necessary to test this hypothesis . The trap design reported here was originally designed and optimized in areas endemic for the savanna dwelling species of S . damnosum s . l . , S . damnosum s . s . and S . sirbanum [10] . It is possible that this design will not prove as effective in attracting other sibling species of the S . damnosum s . l . sibling species , such as those endemic to the forested or mountainous areas of Africa . Indeed , a recent evaluation of the EWT found that it was not very effective in attracting vector black flies in Tanzania [11] , where S . killibanum and S . nkusi are the predominant vectors [20] . Thus , optimization studies like those reported here may need to be conducted in areas where initial trap performance is not satisfactory . A second limitation to the widespread application of the EWT is that the Tangle Trap glue is not locally available in Africa and needs to be imported . Identifying a locally available replacement for Tangle Trap would make widespread application of the traps more likely , as all other components of the trap are inexpensive and available throughout Africa . Recently , we reported the evaluation of an alternative community-based method of vector control , removal of streamside vegetation along the S . damnosum s . l . breeding sites . This so called “slash and clear” method resulted in dramatic ( >90% ) reductions in biting rates that lasted for up to three months after the intervention [21] . It is possible that the application of the slash and clear process together with the deployment of EWTs in selected areas where biting rates are high might result in an almost complete elimination of vector biting . For example , if the effects of the two methods are merely additive , one might hypothesize that combining slash and clear with selective deployment of the EWTs could result in 94–99% reductions in the S . damnosum s . l . biting rate . Apart from the relief provided to the communities from the biting nuisance , this reduction would be expected to accelerate the elimination of O . volvulus transmission . Furthermore , the EWTs might prove to be a good alternative method of community directed vector control in areas where slash and clear cannot be easily applied ( e . g . , along very large and dangerous rivers or in areas where the breeding sites are too small to be easily targeted by slash and clear ) . Studies investigating the effect of combining EWTs with slash and clear are currently underway . | Onchocerciasis or river blindness is historically one of the most important causes of blindness and skin disease in the developing world . It is caused by infection with the filarial parasite Onchocerca volvulus . The finding that ivermectin was an effective and safe treatment for onchocerciasis and the decision by its manufacturer to donate it to treat this infection spawned the development of programs to eliminate river blindness through mass drug administration of ivermectin to the afflicted populations . This has dramatically reduced the prevalence of onchocerciasis worldwide and has resulted in its near elimination in the Americas . But ivermectin alone will not eliminate river blindness in much of Africa; additional interventions are necessary . Here we report the optimization of a simple trap for the black fly vector of O . volvulus and show that these traps can dramatically reduce vector biting in some settings . Together with other recently developed community directed methods of vector control , these traps may augment the effect of the ivermectin distribution programs , bringing the goal of elimination within reach in much of Africa . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"onchocerca",
"volvulus",
"education",
"helminths",
"sociology",
"tropical",
"diseases",
"geographical",
"locations",
"uganda",
"black",
"flies",
"social",
"sciences",
"parasitic",
"diseases",
"animals",
"onchocerca",
"diptera",
"pharmaceutics",
"drug",
"administration",
"neglected",
"tropical",
"diseases",
"onchocerciasis",
"africa",
"schools",
"insects",
"arthropoda",
"people",
"and",
"places",
"helminth",
"infections",
"eukaryota",
"nematoda",
"biology",
"and",
"life",
"sciences",
"drug",
"therapy",
"organisms"
] | 2019 | Optimization and evaluation of the Esperanza Window Trap to reduce biting rates of Simulium damnosum sensu lato in Northern Uganda |
Tropheryma whipplei is known as the cause of Whipple's disease , but it is also an emerging pathogen , detected in stool , that causes various chronic localized infections without histological digestive involvement and is associated with acute infections , including gastroenteritis and bacteremia . We conducted a study in 2008 and 2009 using 497 non-diarrheic and diarrheic stool samples , 370 saliva samples , 454 sera samples and 105 samples obtained from water samples in two rural Sine-Saloum villages ( Dielmo and Ndiop ) in Senegal . The presence of T . whipplei was investigated by using specific quantitative PCR . Genotyping was performed on positive samples . A serological analysis by western blotting was performed to determine the seroprevalence and to detect seroconversion . Overall , T . whipplei was identified in 31 . 2% of the stool samples ( 139/446 ) and 3 . 5% of the saliva samples ( 13/370 ) obtained from healthy subjects . The carriage in the stool specimens was significantly ( p<10−3 ) higher in children who were between 0 and 4 years old ( 60/80 , 75% ) compared to samples obtained from individuals who were between 5 to 10 years old ( 36/119 , 30 . 2% ) or between 11 and 99 years old ( 43/247 , 17 . 4% ) . The carriage in the stool was also significantly more common ( p = 0 . 015 ) in subjects with diarrhea ( 25/51 , 49% ) . We identified 22 genotypes , 16 of which were new . Only one genotype ( #53 ) was common to both villages . Among the specific genotypes , one ( #52 ) was epidemic in Dielmo ( 15/28 , 53 . 4% , p<10−3 ) and another ( #49 ) in Ndiop ( 27 . 6% , p = 0 . 002 ) . The overall seroprevalence was estimated at 72 . 8% ( 291/400 ) . Seroconversion was detected in 66 . 7% ( 18/27 ) of children for whom PCR became positive in stools between 2008 and 2009 . T . whipplei is a common bacterium in the Sine-Saloum area of rural Senegal that is contracted early in childhood . Epidemic genotypes suggest a human transmission of the bacterium .
Traditionally , Tropheryma whipplei was considered to be a rare bacterium that typically caused the classic form of Whipple's disease , which is characterized by histologically periodic acid-Schiff–stained bacilli in infected small-bowel macrophages [1]–[3] . However , this well-known pathology , described primarily in Caucasian men and very seldomly in people of African origin , represents only one rare clinical manifestation of T . whipplei infection [2] , [4] , [5] . Recent studies have shown that a wide spectrum of infections is caused by T . whipplei [2] , [5]–[9] . The bacterium also causes localized chronic infections without histological digestive involvement , such as endocarditis , spondylodiscitis , meningoencephalitis , uveitis , and pneumonia [2] , [5]–[9] . In addition , T . whipplei DNA was recently found to be highly prevalent ( 15% ) in stool samples obtained from 241 children who were between 2 to 4 years old in France and had gastroenteritis , but it was not detected in a control group of children of the same age without gastroenteritis [10] . T . whipplei DNA has also been detected in stool and saliva specimens obtained from healthy individuals , and its prevalence depends primarily on geographic area [11]–[13] . In Europe , the prevalence of this bacterium in stool samples is estimated to be between 1% and 11% among the healthy general population and between 12% and 26% among sewage workers [14]–[16] . In France , the carrier prevalence of T . whipplei in saliva is estimated to be 0 . 2% in the general population and 2 . 2% among sewage workers [14] , [15] . In a preliminary study of 150 healthy children conducted in 2 villages in Senegal ( Dielmo and Ndiop ) , the prevalence of T . whipplei in stool samples was 30% for children between eight months and two years old and 44% in children between two and ten years old [17] . More recently , a study using 204 blood samples obtained from febrile patients who were negative for malaria in these same villages found T . whipplei DNA in 13 samples ( 6 . 4% ) [17] . Here , we have extended this research to the entire population of these 2 villages to confirm these preliminary results and establish the kinetics of T . whipplei carriage in stools from April 2008 to April 2009 . We also extended the study to include saliva samples from healthy individuals , stool samples from diarrheal patients and additional water samples .
This cohort study was approved by the national ethics committee of Senegal and the local ethics committee of IFR 48 ( agreement number 09–022 , Marseille , France ) . Written informed consent was obtained from all individuals , including patients and parents or legal guardians of all children . From April to October 2009 , we performed studies among the populations of Dielmo ( 13°43′N , 16°25W ) and Ndiop ( 14° 33′ N , 16°15′ W ) , which are two villages in the Sine-Saloum region of Senegal . These villages are included in the Dielmo project , a longitudinal prospective study initiated in 1990 for the long-term investigation of host–parasite associations [18] , [19] . In April 2009 , at the time of the samplings of saliva and stools specimens , the population of Dielmo was composed of 379 people ( 200 females ) , including 63 children ( 17% ) of less than 5 years of age and the population of Ndiop was composed of 274 ( 154 females ) , including 55 children of less than 5 years of age ( 20% ) . A total of 105 samples , including 6 samples obtained from wells ( 145 ml of water from each one ) , 92 samples obtained from canaris ( vases used to store water; 50 ml of water from each one ) , and 7 samples from opens water sources in the region ( rivers and marshes; 200 ml of water from each one ) were analyzed . These samples were stored at room temperature for transport to our laboratory in France . On arrival , the water samples were filtered through a 0 . 22 µm membrane ( Millipore ) . The membrane was then immersed in 2 ml of PAS buffer ( Page's Amoeba Saline , Unipath Ltd . Wade Road , UK ) , and the supernatant was stored at −80°C until analysis . Approximately one gram of stool , 200 µl of saliva and 200 µl of water were individually submitted for DNA extraction using a BioRobot MDx workstation ( QIAGEN , Valencia , CA , USA ) according to the manufacturer's recommendations and protocols . Quantitative real-time PCR ( qPCR ) targeting repeated sequences ( repeat-PCR ) was performed using a LightCycler® instrument ( Roche Diagnostics , Meylan , France ) with a QuantiTect Probe PCR Kit as described previously [15] , [20] . First , specimens were tested using the Twhi3F ( 5′-TTGTGTATTTGGTATTAGATGAAACAG-3′ ) and Twhi3R ( 5′-CCCTACAATATGAAACAGCCTTTG-3′ ) primer pair and the specific TaqMan probe Twhi3 ( 6-FAM-GGGATAGAGCAGGAGGTGTCTGTCTGG-TAMRA ) . When the specimen was found to be positive using this assay , the result was confirmed through a second round of qPCR analysis using the Twhi2F ( 5′-TGAGGATGTATCTGTGTATGGGACA-3′ ) and Twhi2R ( 5′-TCCTGTTACAAGCAGTACAAAACAAA-3′ ) primer set and the Twhi2 probe ( 6-FAM-GAGAGATGGGGTGCAGGACAGGG-TAMRA ) . To validate the test , we used positive and negative controls as previously reported [15] , [20] . Genotyping of T . whipplei was performed as described previously [21] . Each of the four highly variable genomic sequences ( HVGSs ) obtained from each specimen were compared with those available in both the GenBank database and our internal laboratory database to determine their corresponding genotype . When a new sequence was obtained , we have systematically performed two additional rounds of sequencing to confirm it . The sensitivity levels of repeat-PCR and PCR assays used for genotyping were evaluated on DNA extracted from 10-fold dilutions of a suspension of 104 T . whipplei strain Marseille-Twist ( ATCC VR-1528 ) bacterium . Finally , as we have not previously amplified T . whipplei in water , we have checked the feasibility to amplify it from water artificially infected . Thus , we have infected 5 flasks of 50 ml of sterile water with 10-fold dilutions of a suspension of 104 T . whipplei . The water samples were then filtered and submitted to DNA extraction and repeat-PCR as reported above . Serological assays were performed by western blot . Native and deglyclosylated samples obtained from total bacterial extracts were prepared for SDS-PAGE as previously reported [22] . Proteins were separated by SDS-PAGE and transferred onto nitrocellulose membranes . The protein concentration of the samples was determined using a Biorad reagent ( Hercules , CA , USA ) . The membranes were immersed in PBS supplemented with 0 . 2% Tween 20 and 5% non-fat dry milk ( blocking buffer ) for 1 h at room temperature before incubation with primary sera ( diluted 1∶1 , 000 in blocking buffer ) for 1 h at room temperature . The membranes were then washed three times with PBS-Tween 20 , and immunoreactive spots were detected by incubating the membranes for 1 h at room temperature with a peroxidase-conjugated goat anti-human antibody ( Southern Biotech , Birmingham , AL , USA ) diluted 1∶1 , 000 in the blocking buffer . Detection was performed using chemiluminescence ( Enhanced Chemiluminescence Western Blotting Analysis System; Amersham Biosciences , Uppsala , Sweden ) with an automated film processor ( Hyperprocessor; GE Healthcare ) . Data were analyzed using PASW statistics 17 software ( SPSS , Chicago , IL , USA ) . Non-parametric values were compared using the chi-square test . Statistical significance was defined as p<0 . 05 . The corrected chi-squared test or the Fisher's exact test was used where indicated .
In this study , stool samples from 446 individuals aged from 1 month to 87 years old ( mean age 20±18 . 71 years ) were collected in April 2009 . Among these samples , 139 ( 31 . 2% , 95% confidence interval [CI] 26 . 9%–35 . 6% ) presented a positive PCR in Dielmo and Ndiop . Sixty-two out of 219 tested in Dielmo ( 28 . 3% ) were positive , compared to 77 out of 227 in Ndiop ( 33 . 9% , p = 0 . 2 ) . T . whipplei carriage in stools was higher in children aged from 0 to 4 years ( 60 out of 80 , 75% , Figure 1 ) . The prevalence in this age group was significantly different from other age groups ( p<10−3; 60/80 versus 36/119 for the group from 5 to 10 years old and 43/247 for the group from 11 to 87 years old ) . Saliva samples were collected from 370 people aged from 4 to 91 years ( mean age 27±19 . 5 years ) . T . whipplei carriage was found to be 3 . 5% ( 13/370 , 95% CI 1 . 9%–5 . 8% , Figure 2 ) in these samples , and the prevalence in Ndiop ( 5 . 9% , 12/201 ) was significantly different from that detected in Dielmo 0 . 6% ( 1/169 , p = 0 . 005 ) . T . whipplei carriage in saliva was higher in the 5 to 10 years old age group ( 8 out of 83 , 9 . 6% , p = 0 . 02 ) . Among the 150 children tested in 2008 , stool samples were obtained from 118 in 2009 . Of 118 , 52 ( 44 . 1% ) were negative over the two-year period . A total of 30 of 118 ( 25 . 4% ) were positive in April 2008 and 2009 . Twenty-eight children whose stools were negative in April 2008 were determined to be positive in April 2009 ( 23 . 7% ) . Stool and saliva samples were available for 294 persons in April 2009 . Of these sample sets , 219 were negative in both stool and saliva ( 74 . 5% ) ; 62 ( 21 . 1% ) were positive in stool but negative in saliva , 13 were positive in both samples ( 4 . 4% ) and none was positive in saliva and negative in stools . The villages of Dielmo and Ndiop contain 82 households . In this study , we analyzed samples obtained from individuals residing in 58 different households . In 52 of the households , none of the residents presented T . whipplei DNA in their saliva samples; among these individuals , 126 of 431 ( 29 . 2% ) were positive for T . whipplei in stool specimens . In the 6 remaining households , at least one of the residents presented T . whipplei DNA in their saliva sample; among those living in these households , 48 of 147 ( 32% ) were positive for T . whipplei in stool specimens . Finally , T . whipplei DNA was not detected in the 105 water samples tested even if we were able to amplify T . whipplei among all the flasks artificially infected . Fifty-one diarrheal stool samples ( Figure 3 ) were analyzed during our study ( mean age 4±3 . 2 years ) . T . whipplei was detected in 10 of 20 samples in Dielmo ( 50% ) and 15 of 31 samples in Ndiop ( 48 . 4% ) . The prevalence of T . whipplei in patients with diarrhea , 49% ( 25/51 , 95% CI 35 . 2%–62 . 6% ) , was significantly higher ( p = 0 . 015 ) than that found among controls ( 139/446 , 31 . 2% ) . When we analyzed our data according to age , our primary observation was that T . whipplei DNA was never detected in diarrheic patients who were more than 11 years old , whereas it was detected in half of the patients under 11 years old . Genotyping data were available when high DNA loads were found , which occurred in 61 specimens ( 53 stool and 8 saliva samples ) from 57 people ( 28 in Dielmo and 29 in Ndiop , Figure 4 and Table S1 ) . Indeed , in comparisons of the detection capacities of the molecular assays , we were able to detect 1 DNA copy of standard control DNA when our repeat-PCR was used and only 10 copies when our PCR assays for genotyping were used . The additional rounds of sequencing have allowed to obtain the same sequence each time confirming the robustness of our data . All the nucleotide sequences detected in this study have been deposited in GenBank and their reference numbers are presented in Table S2 . Overall , 22 different genotypes were detected , including 16 new genotypes ( genotypes 63–81 ) ; all of these were specific to Senegal . Only one genotype ( #53 ) was common to the two villages , although it was identified more frequently in Ndiop ( 7/9 , 77 . 8% ) than in Dielmo ( 2/9 , 22 . 2%; p = 0 . 1 ) . Otherwise , 26 of 28 individuals from Dielmo ( 93% ) and 22 of 29 from Ndiop ( 76% ) were infected with genotypes specific to each village ( Figures 5 and 6 ) . Genotype 52 was observed specifically in 15 of 28 individuals tested in Dielmo ( 53 . 4% versus 0 in Ndiop , p<10−3 ) . Genotype 49 was present exclusively in 8 of the 29 individuals tested in Ndiop ( 27 . 6% versus 0 in Dielmo , p = 0 . 002 ) . For 4 people out of 57 , a genotype was obtained for both saliva and stools specimens . The same genotype was identified in both samples for the 4 individuals . A concordance study was conducted between the presence of T . whipplei in stool among children younger than 11 years old and the presence of an immune response against T . whipplei as determined by western blot analysis . We identified a link between the presence of T . whipplei in the stool and the presence of an immune response against the bacterium . Among the ≤6-years old age group , no patients who presented negative PCR results exhibited a serological response ( Table 1 ) . However , in the 5- to 10-year-old age group , 7 children who presented negative stool-sample PCR results exhibited an immune response against T . whipplei .
The validity of the reported data is based on strict experimental procedures and controls , including positive and negative controls used to validate the test . Each positive PCR result was confirmed with the successful amplification of an additional DNA sequence , and all sequences with at least one mutation were systematically confirmed through 2 additional rounds of sequencing . Therefore , we are confident in the results presented here: T . whipplei is endemic . Although twenty-two different genotypes that are specific to Senegal have been detected , confirming the genetic heterogeneity of T . whipplei [10] , [13] , [21] , only genotype 53 was common to both villages . This genotype was epidemic in Ndiop , where it affected 24% of the positive individuals [13] . All of the other detected genotypes were specific to each village and were endemic . Among them , one genotype ( #49 ) was detected in 27 . 6% of the affected individuals in Ndiop and another ( #52 ) was detected in 53 . 6% of the affected individuals in Dielmo . Thus , the fact that almost half of the people were affected by the same genotype cannot be attributed to chance . This important circulation of specific genotypes confirms the theory that T . whipplei is contagious . One of our initial hypotheses was that the water , mainly the stored water for drinking represented a possible source of contamination for the population; however , our results allowed us to refute this hypothesis [13] . The lack of detection of T . whipplei in our environmental samples may be explained by the fact that toilets in each household are constructed by the principle of septic tank , so the excrements are not allowed to be freely distributed all over . Villagers used as a drinking water relatively deep ( >20 m ) covered wells only , so the contact of the drinking water with the excrements is minimized . Besides , the fact that T . whipplei is a fastidious bacterium , suggests that it cannot propagate in water . Thus , even if minuscule quantities of bacterium may invade water with human excrements , they are , probably , below the threshold of identification and do not play important epidemiological role . Moreover , environmental sources do not explain the circulation in the two villages . Our hypothesis is that the bacterium can be transmitted through saliva [23] . In France , T . whipplei has recently been identified as an agent of gastroenteritis in young children , either alone or in combination with other pathogens [10] . In parallel , an in vivo model of infection with T . whipplei in mice was developed that further confirmed the role of T . whipplei as an agent of gastroenteritis [24] . In our study , the analysis of children who were between 5 and 10 years old confirms a link between T . whipplei and gastroenteritis , although the interpretation of the results from the 4-year-old and younger group is more difficult . The high prevalence of carriage of T . whipplei in young subjects , the small number of diarrheal stool samples analyzed and the occurrence of very early primary infections are factors that limit the interpretation of our data . Nevertheless , it is important to emphasize that the spectrum of the manifestations of primary infection due to T . whipplei seems to be variable and includes gastroenteritis , pneumonia and bacteremia [10] , [17] , [25] , [26] . Furthermore , patients may also develop multiple infections , including diarrhea or successive infections . It will be necessary to design a specific study in order to better determine the prevalence of T . whipplei among young children with diarrhea in rural Senegal as well as those of other defined pathogens and to evaluate the percentage of co-infection . The prevalence of T . whipplei in stool specimens obtained from children under 5 years of age is 75% , a percentage that decreases among the 5 to 10-year-old age group ( 30% ) . The prevalence in individuals older than 10 years old ( 17 . 4% ) is lower than that in children under 10 years old , although it is higher than that observed among the general population of Europe [14] , [15] . A discrepant prevalence of T . whipplei in saliva is observed between the 2 villages . Discrepancies between these 2 villages have been previously observed regarding the incidence of several infectious diseases such as flea-borne spotted fever , tick-borne relapsing fever , malaria , and Q fever that are more prevalent in Dielmo than in Ndiop [27] . However , reasons for the significantly different prevalence of these infectious diseases in the 2 geographically close villages remain unexplained until now . In our study , a higher prevalence of T . whipplei carriage in saliva is observed in Ndiop in comparison to Dielmo . We have no explanation for this . One hypothesis may be a different lifestyle between the 2 populations , with closer contacts between people in Ndiop in comparison to Dielmo but this suggestion as well as other hypotheses should be studied . This is the first study on the seroprevalence of T . whipplei performed in Africa using a recently described methodology [22] . This western blot-based approach revealed that subjects in France who were healthy carriers of T . whipplei exhibited a more intense immune response compared to those with classic Whipple's disease [22] . In our study , one of the 2 children under 5 years old tested positive . For the children between 5 and 10 years old , the seroprevalence is 79% , and it is 71 . 2% for people older than 10 years old . These data suggest that almost two thirds of the population in the Sine-Saloum area of rural Senegal have been infected with T . whipplei , confirming that this bacterium is common . Very few sera from very young children have been tested . However , notably for children younger than 6 years old , the presence of a serological response was systematically associated with the presence of T . whipplei DNA in stool specimens , whereas for children between 5 and 10 years old , almost half who presented a serological response did not present T . whipplei DNA in their stool specimens . This suggests a primary infection that occurs before individuals are 5 years old and a later elimination of the bacterium . Taken together , these seroprevalence and seroconversion data , in addition to the high prevalence of T . whipplei in stool samples from young children , are strong arguments supporting the idea of primary infections occurring in young children . Finally , there is also a network of circulation of epidemic genotypes between households in each village , as well as the 3 epidemic genotypes that are specific for 2 households in Ndiop . Overall , T . whipplei is an emerging pathogen since the first culture of the bacterium 10 years-ago has allowed the development of efficient molecular tools leading to its more common detection . The first data have been obtained in France with approximately 2% of positive in stools from the general population but more impressive are the data observed in this area of rural Senegal with 31 . 2% of positivity in stools . The high incidence of T . whipplei in rural Senegal and the increasing spectrum of clinical manifestations due to the bacterium allow us to suggest that T . whipplei infection might be a major public health concern in West Africa . The existence of epidemic genotypes and its absence from environmental samplings suggests a human transmission of the bacterium . We speculate that T . whipplei is a contagious bacterium that is contracted early in childhood and is responsible for an underestimated number of acute clinical manifestations . Only a small number of individuals with specific immune deficiencies , which have not yet been determined , will develop classic Whipple's disease or other chronic localized infections . Further studies in Senegal will help us to elucidate the natural history of T . whipplei . | Tropheryma whipplei is known as the cause of Whipple's disease . It is also an emerging pathogen , detected in stool that causes various chronic localized infections without histological digestive involvement and is associated with acute infections , including gastroenteritis and bacteremia . We have studied the presence of T . whipplei on non-diarrheic and diarrheic stool samples , saliva samples , and sera samples in two rural Sine-Saloum villages ( Dielmo and Ndiop ) in Senegal . T . whipplei was identified in 31 . 2% of the stool samples and 3 . 5% of the saliva samples from healthy subjects . The carriage in the stool specimens was higher in children who were between 0 and 4 years old ( 75% ) compared to samples obtained from individuals between 5 to 10 ( 30 . 2% ) or between 11 and 99 ( 17 . 4% ) . The carriage in the stool was also more common in subjects with diarrhea ( 49% ) . We identified 22 different genotypes of T . whipplei . Only one genotype was common to both villages . Among the specific genotypes , one was epidemic in Dielmo and another in Ndiop . The seroprevalence was estimated at 72 . 8% . T . whipplei is a common bacterium in the Sine-Saloum area of rural Senegal that is contracted early in childhood . Epidemic genotypes suggest a human transmission of the bacterium . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"bacterial",
"diseases",
"infectious",
"diseases",
"neglected",
"tropical",
"diseases"
] | 2011 | Tropheryma whipplei: A Common Bacterium in Rural Senegal |
Babesia microti is tick-borne disease that is an emerging threat to public health due to increasing prevalence and expanding geographic range . Detection and constant surveillance of babesiosis is imperative for predicting pathogen expansion . Leveraging our whole genome sequence ( WGS ) analyses of B . microti , we developed a single nucleotide polymorphism ( SNP ) -based high resolution melt ( HRM ) surveillance tool . We developed our HRM assay using available sequence data and identified 775 SNPs . From these candidate SNPs , we developed a 32-SNP barcode that is robust and differentiates geographically distinct populations; it contains SNPs that are putatively neutral , located in nuclear , mitochondrial , and apicoplastal regions . The assays are reproducible and robust , requiring a small quantity of DNA ( limit of detection as low as 10 pg . ) . We analyzed the performance of our HRM assay using 26 B . microti clinical samples used in our WGS study from babesiosis endemic regions in the United States . We identified a minimal barcode consisting of 25 SNPs that differentiate geographically distinct populations across all clinical samples evaluated ( average minor allele frequency > 0 . 22 ) . Supporting our previous WGS findings , our 25-SNP barcode identified distinct barcode signatures that segregate B . microti into two lineages: Northeast and Midwest , with the Northeast having three subpopulations: Connecticut/Rhode Island , Nantucket , and the R1 reference group . Our 25-SNP HRM barcode provides a robust means genetic marker set that will aid in tracking the increasing incidence and expanding geographic range of B . microti infections .
Human babesiosis is an increasingly recognized tick-borne disease with the vast majority of cases in the United States caused by the malaria-like protozoan Babesia microti [1 , 2] . Infections occur predominantly through the bite of an infected Ixodes scapularis tick but can also occur via the transfusion of blood products [3 , 4] . Human babesiosis , like malaria , is characterized by fever and haemolysis [3] . Severity of infection ranges from asymptomatic to severe and can be complicated by hemolytic anemia , and respiratory or organ failure [5] . Since the first reported case of human babesiosis in the United States on Nantucket in 1969 , the geographic range and incidence of babesiosis has increased [3 , 6–8] . Human infections primarily occur in the United States ( U . S . ) , particularly in New England , New York , and North Central Midwest [2–5 , 6 , 7] . This led the Center for Disease Control and Prevention ( CDC ) in 2011 to classify human babesiosis as an emerging and nationally notifiable disease [9] . In 2014 , the CDC reported babesiosis cases rose from 1 , 126 to 1 , 744 with 94% of the cases in Connecticut , Massachusetts , Minnesota , New Jersey , New York , Rhode Island and Wisconsin ( https://www . cdc . gov/parasites/babesiosis/data-statistics/index . html ) . The increasing number of endemic areas of B . microti signals the need to effectively track and record disease incidence . In this study , we leveraged our whole genome sequence ( WGS ) analyses of B . microti [1] and developed a High Resolution Melt ( HRM ) [10] single nucleotide polymorphism ( SNP ) genotyping method . Previously , we have developed SNP barcodes for to both Plasmodium vivax and P . falciparum and have employed them to successfully monitor malaria transmission using HRM analysis [11 , 12] . Leveraging our experience , we designed a 32-SNP barcode for B . microti that contains SNPs that are located in nuclear , mitochondrial , and apicoplastal regions . We further reduced this set to minimal 25-SNP barcode for B . microti that can uniquely identify the geographic origin of the sample and may ultimately provide genomic insight for various population parameters . The 25-SNP barcode was evaluated in a pilot screen of 26 clinical samples from babesiosis endemic regions in the continental U . S . : Mainland New England ( MNE—Massachusetts , Maine , Connecticut , Rhode Island and New Hampshire ) , Midwest ( MW—Wisconsin , Minnesota , and North Dakota ) and Cape Cod ( NAN—Nantucket ) . The development of this 25-SNP barcode for B . microti serves to enhance epidemiological surveillance of this emerging pathogen .
We used B . microti strains that were sequenced at the Broad Institute as internal controls to identify both reference and alternate alleles for each assay [1] . The reference alleles were based on version 1 of the sequencing information available in Genbank [13] . The original R1 reference ( REF ) contained three chromosomes , where chromosome 3 was a supercontig ( GI: 399217317 ) . In the revised version , the supercontig was broken to two chromosomes , chromosomes 3 ( GI: 908660426 ) and 4 ( GI: 908661396 ) , for a total of four nuclear chromosomes . For barcode development , we used version 1 for the original R1 reference . Every SNP in our barcode can be mapped to a position in the revised model . We used 26 clinical samples for validation of our B . microti barcode assays . As previously described by Lemieux and colleagues [1] , study participants included patients with a positive Babesia smear or B . microti PCR test . The samples were collected at study sites in four regions in the continental United States: MNE [Massachusetts ( 8 samples ) , Maine ( 1 ) , Connecticut ( 1 ) , and New Hampshire ( 1 ) ] , MW ( Wisconsin ( 1 ) , Minnesota ( 2 ) and North Dakota ( 1 ) ] and NAN [Nantucket ( 7 ) ] and unknown origin ( 4 ) . Venous blood draws of 5–10ml was collected in a heparin tube at the time of enrollment and stored at 4°C for less than 3 weeks . We obtained clinical samples through the Massachusetts General Hospital ( MGH ) microbiology laboratory , Brigham and Women's Hospital ( BWH ) , and at the University of Massachusetts Medical School ( UMMS ) hospital clinical laboratory at UMMS . As previously described by Lemieux and colleagues [1] , informed consent was obtained for all study participants at MGH and BWH according to Partners Institution Review Board ( IRB ) protocol 2014P000948 . Samples obtained from the UMMS were IRB-exempted de-identified discarded peripheral blood samples . We obtained historical samples from The Mayo Clinic as discarded , de-identified specimens . Finally , we obtained historical laboratory-adapted strains and tick samples from Tufts School of Veterinary Medicine , including serial isolates of the GI strain of a recurrent I . scapularis ( tick ) and rodent passage from 1986 to 2014 and the RMNS strain . Genomic DNA was previously harvested from whole blood samples from babesiosis positive ( i . e , diagnosed by BSA ) patients in heparinized tubes . Next , the genomic DNA was extracted from the samples using the QIAamp DNA Blood Mini Kit ( QIAGEN ) . The concentration of DNA for all clinical samples was quantified using the Qubit 3 . 0 Fluorometer ( Invitrogen ) . Next , whole genome amplification ( WGA ) was performed on the clinical samples using the illustra Ready-To-Go Whole GenomiPhi V3 DNA Amplification Kit ( GE Healthcare Life Sciences ) according to manufacturer’s instructions . For each WGA reaction , 2 μl of each clinical sample was used to yield 40 μl amplified product . Following WGA , the samples’ DNA concentration was quantified using the Qubit 3 . 0 Fluorometer . We developed a specific B . microti qPCR assay for research use to detect and quantitate our babesiosis samples . The B . microti qPCR assay targets the cytochrome oxidase B , CytB gene ( NCBI ID: 13435122 ) . We scanned the B . microti genome ( NCBI ID: 11700 ) using Geneious software ( version 6 . 1 ) ( Biomatters Ltd . ) . We identified a 118 base pair region in the CytB gene that was highly conserved . We used BLAST genomic database search ( National Library of Medicine ( NLM ) to confirm that the region was specific to known pathogen microti strains ( RI , Gray and MN-1 ) . We designed primers and probes using Primer3 [14] and Realtime PCR Tool ( Integrated DNA Technologies , IDT ) . We checked primer pairs for potential primer dimers and their specificity using Geneious software ( version 6 . 1 ) , and we used BLAST to screen for species specificity . We identified optimized primer sequences for the forward primer as CCTAGGTATGTATCATCTTAACCTCTTT and the reverse primer sequence as TAGGGATCGTAGTCGTGTACTG . We used PrimeTime Probes ( Integrated DNA Technologies , IDT ) with the fluorescent dye FAM on the 5’end is double-quenched in the center by ZEN and by on its 3’end of the probe sequence ( CCCAAGTAGGTATCTATGTACTTCTACTGT ) . We performed the qPCR assay using the LightCycler 96 System ( Roche ) . The reaction consisted of 5 . 0 μl LightCycler FastStart Essential DNA Probes Master Mix ( Roche ) , 2 . 0 μl of 0 . 1 μM forward and reverse primer mixture , ( 2 . 0 μl ) 0 . 1 μM probe , and 1 μl of DNA sample in a total reaction volume of 10 μl . The 2-step amplification cycling conditions were as follows: 95°C for 10 min and 40 cycles at 95°C for 15 secs , and 60°C for 60 secs . We determined the limit of detection ( LoD ) for the qPCR assay by selecting the lowest gDNA concentration which was able to reproducibly , in triplicate , yield a cycle threshold ≤ 36 cycles . We then ran 20 reactions with template gDNA at that observed LoD . We defined the LoD criteria as having a confidence interval ( CI ) of ≥ 90% . If these criteria were not met , we would repeat the assay at a higher concentration . We identified candidate SNPs for our B . microti barcode by using the UnifiedGenotyper tool from the Genome Analysis ToolKit ( GATK ) [15] with minimum quality score of 50 [1] . We further filtered the identified SNPs by performing principal component analysis ( PCA ) using the R package ‘stats’ [16] . We selected SNPs with a PCA component absolute score ( absolute value of the transformed variable values corresponding to a particular data point ) greater than 3 , 2 , then 0 . 5 . The principal components score represents the distance from the origin in the transformed space , so the greater the absolute value of the principal components score , the more informative the SNP is at that position . We then screened these candidate SNPs by selecting for putatively neutral mutations ( intergenic , intronic , and 4-fold degenerate sites ) and for mutations that resulted class I and II SNPs mutations ( A>G , A>C , T>G , T>C , G>A , G>T , C>A , C>T ) . Finally , we used PCA to identify SNPs that visually create geographic clustering . We assessed the significance of PCA results graphically and by using percent of variance ( POV ) , which is calculated as the sum of eigenvalues corresponding to the first and second principal components over the sum of all eigenvalues [17] . We designed primer pairs using the LightScanner primer design software version 2 . 0 ( BioFire Defense , USA ) . The software uses standard design parameters for HRM , such as primer length ( 18–28 nucleotides ) , melting temperature ( 58–60°C ) , GC content ( 40–60% ) and amplicon size ( <60 bp ) to design optimal primer pairs for each SNP . We then tested the designed primer pairs in silico using uMelt [18] , a flexible web-based tool for predicting DNA melting curves and denaturation profiles of PCR products . Based on the results derived from uMelt , we selected assays with a temperature melt ( Tm ) separation of ≥ 0 . 8°C between the reference and the alternate SNP allele . To ensure the specificity of our primers , we ran a BLAST genomic database search ( National Library of Medicine ( NLM ) , USA ) to compare each primer and probe sequence against common pathogens . We identified the optimal PCR profile in clinical samples as a two-step protocol . The PCR profile involved 120 sec at 95°C; 40 cycles of 30 sec at 95°C and 60 sec at 60°C; and a final HRM cycle of 15 sec at 55°C and 15 sec at 95°C . The optimized master mix for the PCR 12 μl reaction contained 1 μl of DNA sample containing 10 ng/μl DNA in 1X TE Buffer , 1 . 2 μL of PCR grade water ( VWR , Radnor , PA , USA ) , 4 . 8 μL of 2 . 5X LightScanner master mix ( BioFire Diagnostics Inc . , Salt Lake City , Utah , USA ) , and 5 μL of primer solution containing 0 . 1 to 0 . 5 μM of forward and 0 . 1 to 0 . 5 μM reverse primers diluted in 1X TE Buffer depending on the individual assay ( Integrated DNA Technologies ) for a total reaction volume of 12 μL . All assays were performed on an Applied Biosystems ViiA 7 Real-Time PCR System or QuantStudio 6 ( Life Technologies ) . To optimize the HRM assays , we evaluated five different primer concentrations ( 0 . 1 μM , 0 . 2 μM , 0 . 3 μM , 0 . 4 μM , and 0 . 5 μM ) while leaving all other reaction conditions unchanged . We selected primer concentrations based on the correct amplicon size , the cycle threshold ( CT ) ≤ 30 , having a single melt profile , and being robust with an efficiency between 90–100% . We defined a successful assay as having a CT ≤ 30 , because cycle thresholds greater than 30 can result in shifted melt profiles , resulting in unreliable genotyping data [10] . We used the HRM method to genotype our SNPs following the defined protocol . To perform the HRM assay , we first quantified the concentration of DNA for all clinical samples , as described above using a Nanodrop . We then diluted all clinical samples using 1x TE Buffer to 10 ng/μl based on OD260 . For all assays , we included sequenced control samples to identify the reference and alternate allele SNP temperature melt ( Tm ) curves . Next , we prepared a reaction master mix , consisting of 1 . 2 μL of PCR grade water ( VWR , Radnor , PA , USA ) , 4 . 8 μL of 2 . 5X LightScanner master mix ( BioFire Diagnostics Inc . , Salt Lake City , Utah , USA ) , and 5 μL of primer solution containing 0 . 1 to 0 . 5 μM of forward and 0 . 1 to 0 . 5 μM reverse primers diluted in 1X TE Buffer depending on the individual assay ( Integrated DNA Technologies , Inc . ) . We added 1 μl of diluted DNA sample ( 10 ng/μl ) to this master mix . Once we prepared our plate , we gently centrifuged them before beginning PCR , using the profile described above . We evaluated assay efficiency using the standard curve method for qPCR . We prepared the DNA sample , that was previously amplified using WGA , by using six 10-fold serial dilutions ( 103 ng/μl to 1 ng/μl ) of our clinical control sample ( Gray ) and Reference 1 ( REF ) . We generated the standard curve from the qPCR assay to calculate the percent efficiency of each assay and to determine their sensitivity or the minimum amount of DNA needed for successful amplification . We evaluated the B . microti qPCR and SNP barcode assays for analytical specificity in silico by comparing each primer and probe sequence against those of other common causes of pathogenic illness in humans including other protozoan species including: Plasmodium ( P . falciparum , P . vivax , P . knowlesi , P . malariae , P . ovale ) , Babesia ( B . divergens , B . divergens-like , B . venatorum , B . rodhaini , B . bovis ) , and Theileria parva . We used BLAST to evaluate all primers and probes for cross species reactivity using all possible combinations . We tested control samples for each assay and used the derivative Tm curve to identify the reference and alternate alleles for each SNP assay . We evaluated genotyping reproducibility by performing each SNP assay in duplicate . We then calculated the mean Tm differences and standard error between duplicates to evaluate reproducibility and robustness of each assay . Additionally , we further tested genotyping accuracy of our barcode assays on 26 clinical samples . For these clinical samples , we compared HRM SNP data to WGS SNP calling to reaffirm the genotyping reproducibility of our method . We calculated the minor allele frequency ( MAF ) from allele counts for each SNP in each population . We assessed sample uniqueness by comparing genotypes across all pairs of samples . We classified pairs of distinct monomorphic genotypes ( e . g . A/G ) . We performed PCA using the program SmartPCA in the Eigensoft package [17 , 19] on the diverse panel of 26 B . microti clinical samples . We calculated the POV explained as the sum of eigenvalues corresponding to the first and second principal components over the sum of all eigenvalues .
We evaluated the robustness and efficiency of the B . microti qPCR assay by performing the standard curve method using a 10-fold dilution series of the DNA . We performed this experiment in triplicate and had a resultant efficiency of 99 . 4% ( R2 = 0 . 99 ) . Based on the standard curve method , we identified the tentative LoD as 3 pg . We confirmed the LoD by testing 20 replicates at 3 pg and found 19 out of 20 samples were detected ( CI = 95% ) with an average PCR cycle threshold ( CT ) of 34 . 68 . We used BLAST to evaluate assay specificity and potential cross-reacting target sequences and did not identify any significant sequence matches . We evaluated reproducibility and performance of the B . microti qPCR assay by performing a pilot screen in duplicate using our 26 clinical samples , that were previously amplified by WGA , from MNE , MW and Cape Cod that were previously confirmed positive in our WGS study [1] and one sample of Lyme borreliosis as a negative control . We identified 26 out of 26 B . microti samples with an average standard of error of ± 0 . 16 and found that the L . borreliosis was negative with no amplification ( S1 Table ) . Our selection criteria to identify candidate SNPs for the B . microti SNP barcode included using putatively neutral genomic loci ( intergenic , intronic , or 4-fold degenerate sites ) located in nuclear , mitochondrial , and apicoplastal regions . Using B . microti genome information from our previous sequencing efforts [1] on 32 human babesiosis clinical samples from the Northeast ( NE ) ( MNE , Sandy Neck , and NAN ) and MW ( Minnesota and Wisconsin ) , we identified 2 , 445 candidate SNPs . We screened these candidates using PCA and selected 775 SNPs that captured a high degree of population diversity and that differentiated geographically distinct populations . Further screening these candidates for putatively neutral genomic loci and for class I and class II mutations , which have the broadest SNP melt temperature ( Tm ) separation ( A>G , A>C , T>G , T>C , G>A , G>T , C>A , C>T ) , we identified a final set of 477 candidate SNPs for HRM assay development . Next , we winnowed these candidates down based on HRM primer design guidelines . We used uMelt [18] , a flexible web-based tool for predicting DNA melting curves and denaturation profiles of PCR products , and selected 91 candidates SNPs that had the strongest performing assays in silco . Targeting these 91 SNPs , we designed primers for HRM analysis that were specific to B . microti ( S2 Table ) . We tested these HRM assays for species specificity and the ability to distinguish between alternate and reference alleles , with a temperature separation greater than 0 . 8°C . This yielded a 32-SNP barcode , containing putatively neutral genomic loci and including 17 nuclear , 14 mitochondrial , and 1 apicoplastal SNPs ( S3 Table ) . We evaluated the efficiency and LoD of the assays by performing the standard curve method using 10-fold serial ( 103 ng/μl to 1 ng/μl ) dilutions in triplicate . The efficiency of our assays ranged from 90–100% , and the LoD varied from 1 to 10 pg ( S4 Table ) . We evaluated the reproducibility and performance of our 32-SNP barcode by screening the 26 clinical samples , that were previously amplified by WGA , used to test our B . microti qPCR assay and one L . borreliosis sample , used as a negative control . We ran the pilot screen using the Applied Biosystems ViiA 7 Real-Time PCR System . The assays were highly sensitive and accurate , correctly genotyping all 26 clinical samples ( 832 out of 832 SNP calls and concordant across all duplicates ) . The L . borreliosis sample was negative with no amplification . The assays were robust , with the variability in the Tm values for each assay run in duplicate of < 0 . 12°C for all 32 assays and the average standard error ranging from ≥ ± 0 . 02 to ≤ ± 0 . 20°C ( S5 Table ) . Together , this set of 32 SNPs spans all three nuclear chromosomes , mitochondrial and apicoplastal regions , with the closest pair of nuclear SNPs at least 801 bp apart . On average , nuclear SNP pairs are 166 , 646 bp apart , mitochondrial SNPs are 101 bp apart , and apicoplastal SNPs are 2023 bp apart . Together , this set of 32 SNPs successfully distinguished babesia samples tested from populations in MNE , the MW , and NAN . The barcode distinguished all samples according to geographic origin , except for one sample from North Dakota ( ND11 ) which grouped with the MNE samples , and one sample from South Dennis , MA ( Bab14 ) which grouped with the NAN samples ( S6 Table ) . The SNPs captured high degrees of diversity with the average MAF value > 0 . 22 for 53% of the SNPs ( 17 out of 32 ) ( S7 Table ) . We identified a minimal barcode set creating a 25-SNP barcode maintaining nuclear ( 11 SNPs ) , mitochondrial ( 14 SNPs ) and apicoplastal ( 1 SNP ) regions ( Fig 1 ) . The 25-SNP barcode identified distinct barcode signatures that segregate B . microti into two lineages: NE and MW . The NE lineage has three subpopulations Connecticut/Rhode Island ( CT/RI ) , NAN and the Reference 1 group ( REF ) . ( Fig 2 ) . Within the lineages , the barcodes were identical or clonal , except for the MW lineage Wisconsin which differed from Minnesota by one allele ( Assay 7 ) , and the REF lineage New Hampshire which differed from Connecticut by two alleles ( Assay 9 and Assay 16 ) . The barcode captured deep divergence between CT/RI and MW lineages , with the MNE barcode consisting of 96% major alleles ( 24 out of 25 ) and the MW barcode consisting of 40% major alleles ( 10 out of 25 assays ) . Using PCA we demonstrated that the reduced 25-SNP barcode visually separated samples of different geographic origin , including CT/RI , REF , NAN , and the MW ( POV = 72 . 92 ) , and that it is comparable to PCA using all 2 , 445 SNPs ( POV 87 . 31 ) ( Fig 3 ) .
Here , we present a B . microti 25-SNP barcode , demonstrating that it is a robust surveillance tool that could allow scientists to rapidly track and differentiate a parasites geographic origin . The increasing number of B . microti endemic regions highlights the necessity of such a surveillance tool . The 25 SNPs in the barcode are putatively neutral , with high MAF in geographically diverse regions , and span all nuclear chromosomes as well as the mitochondrial and apicoplastal genomic regions . The combination of SNPs from the nuclear , mitochondrial and apicoplastal genome leverages their distinct inheritance patterns . Apicoplastal and mitochondrial DNA SNPs help broadly differentiate geographically distinct populations , while nuclear SNPs can offer deeper insights into populations structure and genome evolution . The barcode is robust , reproducible , and sensitive , requiring only a small quantity of DNA to produce reliable results with a universal limit of detection of 10 pg . The HRM method is highly sensitive and can detect even a single nucleotide difference . We were able to successfully genotype all SNPs ( 468 out of 468 SNP calls ) in all 26 B . microti clinical samples tested in our pilot screen . We demonstrated the potential of the 25-SNP barcode as a surveillance tool in a pilot screen of 26 clinical samples from babesiosis endemic regions in the continental U . S . : MNE , the MW , and NAN . These 25 SNPs were able to accurately identify all samples by geographic origin , except in two cases , one from North Dakota and one from South Dennis , MA . As previously reported in our WGS study , we do not have travel information available to determine if the North Dakota case was imported case . We were , however , able to identify the South Dennis , MA case was potentially locally imported babesiosis from NAN , consistent with the barcode prediction [1] . The 25 SNPS captured high levels of clonality or minimal nucleotide diversity within the populations , and they identified unique barcode signatures that segregate B . microti into four distinct lineages: CT/RI , NAN , REF , and the MW . Further evaluation of the barcode using samples obtained from Connecticut and New Hampshire will determine if the barcode can differentiate these populations . The barcode captured deep divergence between the MNE and MW lineages . All of our findings for the B . microti 25-SNP barcode are concordant with our WGS study [1] as well as recent genotyping studies using variable number tandem repeat ( VNTR ) markers [20 , 21] . The 25-SNP barcode is adaptable , and scientists can further modify this barcode to fit their use . For instance , the completed 32-SNP barcode , provides 17 additional nuclear SNPs that could be evaluated in future population genetics studies . Alternatively , clinical testing the 25-SNP barcode could be further reduced to identify a minimal barcode that captures populations diversity or to create a region-specific barcode . In summary , the B . microti 25-SNP-based barcode can serve as a baseline universal set of assays to distinguish B . microti infections based on geographic origins and to gain insight into changes in parasite population dynamics , transmission , and expansion . Our approach using HRM was strategic , the platform is flexible and can be readily updated . We envision that as research groups use this baseline set of 25-SNPs , their data and findings should be portable across studies , facilitating more accurate comparisons of results and the development of a shorter baseline set or even regional barcodes . Additionally , we envision this method to eventually evolve to potentially create SNP barcodes for additional variants , such as drug resistance , as more information about B . microti becomes available . | Babesia microti is an emerging tick-borne disease and is becoming a public health problem . Over the past two decades , the I . scapularis tick population , which is primarily responsible for human infection , has exploded , doubling the number of babesiosis and other I . scapularis-borne disease cases to approximately 48 thousand reported in 2016 . The increasing number of endemic areas of B . microti signals the need to develop robust and accurate surveillance tools to effectively monitor and record disease incidence . Here , we used our whole genome sequence analysis of B . microti to develop a genetic barcode for B . microti , composed of 25 robust variants . We show the validation and utility of this SNP barcode on 26 babesiosis positive clinical samples from endemic regions in the United States . This genetic barcode provides a means to identify the genetic origin of a parasite and ultimately gain insight into B . microti population structure and transmission dynamics . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"united",
"states",
"medicine",
"and",
"health",
"sciences",
"parasite",
"groups",
"geographical",
"locations",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"parasitology",
"babesiosis",
"organisms",
"north",
"america",
"apicomplexa",
"protozoans",
"genome",
"analysis",
"mitochondria",
"molecular",
"genetics",
"molecular",
"biology",
"techniques",
"bioenergetics",
"genotyping",
"cellular",
"structures",
"and",
"organelles",
"connecticut",
"research",
"and",
"analysis",
"methods",
"genomics",
"artificial",
"gene",
"amplification",
"and",
"extension",
"molecular",
"biology",
"people",
"and",
"places",
"biochemistry",
"eukaryota",
"cell",
"biology",
"babesia",
"genetics",
"biology",
"and",
"life",
"sciences",
"energy-producing",
"organelles",
"computational",
"biology",
"polymerase",
"chain",
"reaction"
] | 2019 | Development of a SNP barcode to genotype Babesia microti infections |
An unprecedented Zika virus epidemic occurred in the Americas during 2015-2016 . The size of the epidemic in conjunction with newly recognized health risks associated with the virus attracted significant attention across the research community . Our study complements several recent studies which have mapped epidemiological elements of Zika , by introducing a newly proposed methodology to simultaneously estimate the contribution of various risk factors for geographic spread resulting in local transmission and to compute the risk of spread ( or re-introductions ) between each pair of regions . The focus of our analysis is on the Americas , where the set of regions includes all countries , overseas territories , and the states of the US . We present a novel application of the Generalized Inverse Infection Model ( GIIM ) . The GIIM model uses real observations from the outbreak and seeks to estimate the risk factors driving transmission . The observations are derived from the dates of reported local transmission of Zika virus in each region , the network structure is defined by the passenger air travel movements between all pairs of regions , and the risk factors considered include regional socioeconomic factors , vector habitat suitability , travel volumes , and epidemiological data . The GIIM relies on a multi-agent based optimization method to estimate the parameters , and utilizes a data driven stochastic-dynamic epidemic model for evaluation . As expected , we found that mosquito abundance , incidence rate at the origin region , and human population density are risk factors for Zika virus transmission and spread . Surprisingly , air passenger volume was less impactful , and the most significant factor was ( a negative relationship with ) the regional gross domestic product ( GDP ) per capita . Our model generates country level exportation and importation risk profiles over the course of the epidemic and provides quantitative estimates for the likelihood of introduced Zika virus resulting in local transmission , between all origin-destination travel pairs in the Americas . Our findings indicate that local vector control , rather than travel restrictions , will be more effective at reducing the risks of Zika virus transmission and establishment . Moreover , the inverse relationship between Zika virus transmission and GDP suggests that Zika cases are more likely to occur in regions where people cannot afford to protect themselves from mosquitoes . The modeling framework is not specific for Zika virus , and could easily be employed for other vector-borne pathogens with sufficient epidemiological and entomological data .
The proposed model is evaluated at a country level , and therefore dependent on country level data for input . The socioeconomic data , epidemiological data , travel data , and vector suitability data for the principle spreading vectors species are all aggregated to the country level for all countries and territories in the Americas as well as the individual U . S . States . Each variable considered in the model is described in further detail below . All data that we can make publicly available are listed in S1 Data .
The inputs for the GIIM consist of the underlying network structure , attributes of both the nodes and links of the network , and observations on the actual transmission process . In this work the network structure is defined by the passenger air travel movements between regions in the Americas . We denote graph G as G ( V , E ) , where VG is a set containing all the vertices of the graph , while EG contains all the edges of the graph . Edges are denoted as eu , v , where u is the source and v is the destination of a travel route represented by the edge . In this paper , VG contains all countries in the Americas , the French , Dutch , British and American overseas territories , and the individual states of the US , which also defines the spatial resolution of the model . The temporal resolution of the model implemented in this work is at a monthly level , which corresponds to the available travel and suitability data . The edges of the graph exists between a pair of regions if there were passenger air travel movements between them during any month in 2015 . The travel patterns between countries is asymmetric , therefore the resulting graph is directed . We denote this graph as GA . We assign a number of attributes to the vertices and edges of the graph to capture the potential risk factors that may influence the regional spread of Zika virus over time . These attributes correspond to the variables described in the data section . The attributes take the form of normalized real values between zero and one , and include the following: A subset of the attributes are dynamic , i . e . their value changes over time , and therefore are denoted with an additional time index , t . The dynamic attributes include the passenger volumes , the incidence rates in each region , and the vector suitability . The monthly airport travel volumes , V u v t , are aggregated to the state and country level . The mean and standard deviation of the vector suitability for each region for each month is computed as described in the data section . The remaining attributes are static , and are assumed not to vary over the course of the year . The GIIM defines a transmission model on the input graph . Graph-based transmission models require a real value w e t ∈ [ 0 , 1 ] , e ∈ E G A to be present on all edges of the graph , these are called edge transmission probabilities . In this application of the GIIM , the attributes are incorporated into a functional form to represent each time-dependent edge transmission probability . The function takes the form as shown below: w e t = C + α V u v t + β I u t + γ S v t + δ S V v t + ω E v + ρ H v + λ P v ( 1 ) The variables in the function correspond to the set of attributes previously listed , and the coefficients of each attribute are the parameters to be estimated by the model . Values w e t are bounded between 0 and 1 . We will denote the surjective assignment of edge transmission probabilities to the edges as W G A : E G A ↦ [ 0 , 1 ] . In addition to the network attributes , GIIM requires a reference observation of the real-life transmission process it seeks to estimate . In the current application , the observation used is the date of the first reported local Zika cases in each region for the time period considered . The reference point is therefore given as a set of 18 binary vectors; each binary vector corresponds to a month of the observed Zika virus outbreak starting from July 2015 to December 2016 , and assigns a value of 0 or 1 to each node of the graph indicating its transmission state . A value of 1 indicates that the presence of local transmission of Zika virus in the region was reported within or before the corresponding month , and a value 0 indicates that Zika cases have yet to be reported from the region . The GIIM relies on an underlying stochastic simulation to model the spreading process . The compartmental model that is used in this paper is the SI model , which has two states: susceptible ( S ) and infected ( I ) . The graph-based SI infection model is an iterative discrete-time model that assigns states to the nodes of the graph: each node can only be in a single state at any time step , and all nodes must have a state at all times . While GIIM can accommodate a more complex compartmental model , e . g . , SEIR , the SI model is selected to fit the Zika virus application based on the assumption that once Zika virus becomes locally established , that region remains a potential risk of furthering the spread of the virus for the time frame being considered . The reason for allowing the option of non-zero outgoing risk after the reported case count reduced to zero is that there could still be infected individuals in the region , especially given the high rate of asymptomatic cases and reporting error . However , it is important to note that the use of the SI model does not enforce a region to have a positive transmission probability over the entire period modeled , it simply allows a non-zero transmission risk value to be estimated by the model . The actual time-dependent transmission probability is defined as a function of the incidence rate at the origin country among other factors; a probability of zero is a feasible solution of the model , and is actually assigned to many of the links over the course of the outbreak . Given the estimated transmission probabilities , in each step of the simulation , “infected” nodes try to “infect” all their susceptible neighbors according to the transmission probability we connecting them . If the attempt is successful , the neighbor will be infected in the following iterations . If the attempt is unsuccessful , the neighbor remains in a susceptible state , and the infected node can continue to make attempts in the following iterations indefinitely . More formally , the transmission process starts from an initially infected set of nodes A 0 ⊂ V G A at iteration t0 . The rest of the nodes V G A \ A 0 are susceptible at the beginning of the process . Let A i ⊆ V G A be the set of infected nodes at iteration i . At each iteration , t , each node u ∈ Ai tries to infect each susceptible neighbor v ∈ V G A \ A i according to the probability w e i , e = ( u , v ) ∈ E G . If the attempt is successful , v becomes infected starting from the following iteration: v ∈ Ai+1 . If more than one node is trying to infect v in the same iteration , the attempts are made independently of each other in an arbitrary order within the same iteration . By definition , the transmission process stops at iteration t if A t = V G A . Since we only simulate a finite portion of the Zika virus epidemic , we stop all transmission models after 18 iterations , corresponding to the observed months of the epidemic at the time of analysis . Within each model run , the transmission process is repeated 10000 times to produce a real value indicating the likelihood of each node being in an infectious states at each iteration , t . The value is calculated by counting the number of repetitions when nodes were in an infectious state , and dividing by the total number of repetitions i . e . 10000 . The GIIM [41] implemented in this work formulates the estimation of edge transmission probabilities as a general optimization task . The model relies on knowledge of the underlying graph and ( at least a subset of ) observations from a transmission process taking place on the network . The real observations take the form o → t ∈ O , where o → is a binary vector and t is a time stamp . Each vector represents a point in time and o → t assigns a binary value to all v ∈ V G A indicating the observed transmission ( or lack there of ) of Zika virus in the region at time t . Set O contains all observations , while set T contains all sample times , i . e . |O| = |T| = 18 . The inputs of GIIM are: an unweighted graph G , a transmission model I , the set of sample times T , and the set of observations O , where O = I n f ( G , W , I , T ) . In this work I is the SI stochastic simulation transmission model defined in the previous subsection , WG: EG ↦ [0 , 1] is the unknown weight assignment , and Inf is a procedure that generates observations at sample times T based on transmission process I taking place on graph G with assigned edge weights we ∈ W , e ∈ E ( G ) . The set of observations , O′ , is a time-dependent vector of real values equal to the probability each node is infected at each timestep , computed from the set of runs . The task of GIIM is to find an estimation W′ of W so that the difference between O and O ′ = I n f ( G , W ′ , I , T ) is minimal . Due to the need to compare a set of binary vectors with a set of real vectors , we compare observations O and O′ using ROC evaluation by pairwise comparing vectors o → i ∈ O and o ′ → i ∈ O ′ for i = 1…18 , computing the AUC value for each pair and averaging over all pairs . The formal definition of the GIIM is as follows: General Inverse Infection Model: Given an unweighted graph G , and transmission model I , the set of sample times T and observations O = I n f ( G , W , I , T ) , we seek the edge transmission probability assignment W′ such that the difference between O and O ′ = I n f ( G , W ′ , I , T ) is minimal . The GIIM defines the estimation of W as an optimization problem , which is solved using an iterative refinement algorithm . The procedure begins with an initial weight configuration W 0 ′ , runs transmission model I , makes observations O′ and computes the error between O and O′ . Based on the error , W′ is refined and the process is repeated , until the error becomes less than an accuracy constant a selected by the user . The search strategy used in this paper is the Particle Swarm Optimization ( PSO ) method [57] . According to the findings in [41] , the method is stable and is able to produce outputs close to the reference , and therefore the solution method chosen for this work as well . It can also be implemented in a parallel environment speeding up computations considerably . Algorithm 1 summarizes the iterative GIIM algorithm . Algorithm 1 Generalized Inverse Infection Model 1: Inputs: G , I , T , O , a 2: Choose initial edge infection probability assignment W′ 3: repeat 4: Compute O ′ = I n f ( G , W ′ , I , T ) 5: Compute d ( O , O′ ) 6: if d ( O , O′ ) ≤a then 7: return W′ 8: else 9: Choose new W′ according to the PSO search strategy . 10: end if A major modification was necessary in order to adapt the GIIM method to be applied in the context proposed here . The original GIIM method estimates the edge transmission values of the graph directly as real values that are static , i . e . , they do not change over time [41] . In this work the edge weights are given as functions of known attributes on the nodes and edges of the graph , and the task becomes the estimation of these functions , or more specifically , the coefficients of these functions . Several attributes in this application are dynamic , i . e . , they change with time . Thus , it is necessary to further extend the function estimation method to account for dynamic attributes , and to adapt the simulation model to handle dynamic edge transmission values . In this work the edge transmission values are defined using a linear function of known attributes , as defined in eq ( 1 ) . More generally , if a i t ( e ) , e ∈ E G A is the set of attributes , where i represents the i-th attribute and t represents the time period , then the time-dependent edge transmission probabilities w e t are given as w e t = g ( f 1 ( a 1 t ( e ) , c 1 → ) , f 2 ( a 2 t ( e ) , c 2 → ) , … , f ℓ ( a ℓ t ( e ) , c ℓ → ) , c g → ) for all e ∈ E G A , where ℓ is the number of available attributes , f1 , … , fℓ and g are functions and c 1 → , … , c ℓ , c g → → are coefficients of functions f1 , … , fℓ , g . Following the notation proposed in [41] , the functions used to compute the edge transmission probabilities are given as a set of attribute functions f1 , … , fℓ assigned to each invididual attribute , an aggregator function g with the role of creating a single value from the result of the attribute functions and a normalization function ensuring that the result falls between 0 and 1 . This formulation makes implementation of the method easy while retaining the flexibility of the model . Let C denote the set of all coefficient vectors . The values of the attributes can change over time , however the functional form and estimated coefficients are assumed to remain constant over the time period examined . The optimization process of GIIM changes from the estimation of the direct weight assignment W to the estimation of C . This provides a means to identify the key factors contributing to the spread of Zika virus throughout the regions in the Americas . A second advantage is technical; |C|<< |W| , therefore the optimization process is much easier because we are only looking for a limited number of function coefficients as opposed to the edge transmission probabilities for all edges of the graph . For the implementation of the model in this paper , all parameters are bounded between -0 . 5 and 0 . 5 , in order to reduce the solution space for the PSO method . Finally , the resulting edge transmission values are trimmed above 1 and below zero by taking w e t = M A X ( 0 , M I N ( 1 , ∑ i = 1 ℓ f ( a i t ( e ) , c i ) + c g ) ) ) .
The first task of our study is to identify the set of attributes ( and corresponding contribution of each ) to be included in the model . We consider the entire set of attributes previously presented in the data section . A linear weighted sum function as defined in eq ( 1 ) is used to compute transmission probabilities , and the dynamic GIIM method is implemented to estimate the coefficients of the function . Different variable configurations were considered , and the model that produced the best fit was selected . The model fit is based on the quantifiable performance metric , ROC AUC averaged over the entire time period . To evaluate and ensure stability of the proposed estimation method , we ran the algorithm 20 times with the same set of inputs and computed the mean and variance of the estimated model coefficients over all runs . The results of the final model are presented in Fig 5 . w e t = 0 . 040 + 0 . 079 V u v t + 0 . 247 I u t + 0 . 174 S v t - 0 . 372 E v + 0 . 285 P v ( 2 ) The results of the model are highly robust . The estimated coefficients vary minimally across runs , and , more significantly , the ranking of all risk factors remained consistent across all runs ( Fig 5A ) . The expected value of the estimated coefficients represent the relative influence of each attribute in the risk of Zika virus spread between a pair of regions . When interpreting these coefficients it is important to first note that this model is designed to estimate the risk of Zika virus spread between two regions resulting in local , vector-borne transmission . This is different from the risk of Zika infected passenger arrivals . For example , there are multiple locations where travel cases of Zika were continually reported , yet no local cases resulted [27] . The lack of local transmission in these examples could be due to many explanations , including insufficient populations of competent vectors and/or intense surveillance and vector control programs implemented at the destination . Thus , a high number of travel-reported cases does not necessarily translate to a high local transmission risk , and for this reason , the risk of Zika infected passenger arrivals and local transmission risk should be modeled separately . Due to the potential harm posed by local outbreaks of Zika , local transmission risk is the primary focus our analyses . The important distinction between modeling travel-reported cases and local transmission risk is perhaps most evident from the low coefficient estimated for travel volume . In fact , we found infected travelers to be significantly less influential than all the other risk factors . This can be explained by the fact that there is a high level of connectivity between most pairs of regions , and more importantly , the highest volume of air travel exists between and within the U . S . states . Yet , with the exception of Florida [24] and Texas [58] , Zika was not broadly established in the U . S . Thus , we do not identify travel volume to be a driving forces in the spread and transmission of Zika virus in the Americas; travel is a necessary , but not sufficient condition . It is worth noting that a model seeking to estimate the risk of infected passenger arrivals alone would likely find this variable to play much more substantial roles . In contrast , we found that Zika virus spread and local transmission is largely driven by regional attributes at the origin ( incidence rate ) and destination: Ae . aegypti suitability , human population density , and the GDP , with GDP being the most significant ( Fig 5A ) . As expected , a higher incidence rate at the travel origin , which can act as a proxy for the likelihood of an infected individual arriving at the destination , significantly increases the risk posed to the travel destination . Similarly , the results indicate a higher vector suitability at the destination corresponds to an increased risk of transmission . High human population density at the destination is also revealed to increase the risk of transmission , consistent with the required presence of both vectors and hosts for mosquito-borne virus transmission . The health indicator variables were excluded from the final model , as they were not found to have a significant impact . Based on our model results , the most dominant and only negative risk factor is a region’s GDP , i . e . , a higher GDP at the destination corresponds to a lower risk of transmission . Based on the magnitude of the coefficient , the destination’s GDP contributes more than any other risk factor considered . The highly negative coefficient of GDP can be explained by the substantial delay ( or complete lack ) of local transmission in the wealthier U . S . states and certain territories and islands in the Caribbean . GDP is obviously not directly involved in Zika virus transmission , but it may indirectly influence the interactions between components of the cycle: hosts and vectors . Poorer nations likely have lower housing qualities and inhabitants may be exposed to more mosquito bites , e . g . , a lack of screens on windows and doors allowing mosquitoes to enter . They may also have more debris around their homes acting as breeding containers for Ae . aegypti . Lastly , GDP may be a proxy for the available surveillance and vector control resources at the destination , an increase of which would aid in reducing local transmission . Our final model captures the relative contribution of both static and dynamic risk factors to explain the spread and local transmission of Zika virus in the Americas . The AUC ROC for this model averaged over the 18 months is 0 . 923 , which indicates an excellent fit ( Fig 5B ) . For the first month we see perfect classification , since the epidemic was only reported in Brazil , which proves trivial for the estimation method . The second half of 2015 corresponds to a major increase in the number of regions reporting local transmission ( as illustrated in Fig 1 ) . Specifically , between November and December the number of regions reporting transmission nearly doubled from 9 to 16 , and then increased to 32 in January . This proved to be the most challenging part of the estimation process , and the reason for the performance drop below 0 . 9 during the period from October to January . The minimum AUC value of 0 . 83 occurred in November 2015 , which is still considered to be good performance for any classifier . The estimation task gets easier again for the last 9 month characterized by a low but constant rate of spreading between the regions , and performance goes around 0 . 95 again for these months . We can conclude , that even though sudden bursts in the reported local outbreaks decrease prediction accuracy slightly , the method is able to provide accurate estimation . The estimation process took between 3 to 4 hours with the number of iterations between 250 and 350 . A parallel version of the algorithm was implemented in C++ , and the results were computed on a PC with an 4-core i7-7700k 4 , 2 GHz processor . Using the final estimated model , we computed the probability of Zika virus spread between any pair of regions in the Americas resulting in subsequent local transmission ( Fig 6 ) . The link probabilities reveal the highest risk travel routes connecting regions at discrete points in time over the outbreak , as well as how the relative ranking changes over time ( Fig 6A ) . Out of the 2946 feasible edges in the network , only 711 edges have nonzero transmission probabilities , and only 58 of the edges has a value greater than 5% at any point of the observation period . The time-dependent data for the top 100 transmission links , exportation risk , and importation risk are provided in S1 Data . In general , the high risk travel routes are dominated by ones outbound from the Caribbean Islands with the highest incidence rates ( earlier in the outbreak ) , and pointing to the less developed countries and territories of the Caribbean and Central America . These results are corroborated by the early estimated introduction times into countries like Haiti and Honduras relative to Puerto Rico , Mexico , and the U . S . [3 , 4 , 24] , see also www . nextstrain . org/zika . The risk on routes departing a given country , e . g . , Brazil , behave similarly , but vary in magnitude . They also display different behavior than the outgoing risk posed by other high risk countries , e . g . , Martinique . The estimated transmission probabilities fluctuate over time due to changes in the dynamic attributes at both the route origins ( outbreak scale ) and destination ( vector suitability ) , as well as variations in the monthly travel volumes between regions . The effect of these risk factors , for example , the lower Zika incidence rates in Puerto Rico relative to Martinique ( Fig 1B ) , is illustrated by the corresponding temporal differences in transmission risk into Haiti ( Fig 6A ) . The route level risk can be aggregated to provide import and export risk profiles at a regional level . This type of spatial and time-dependent information can help guide policy decisions , such as where to allocate available resources at different stages during an outbreak . The regional level risk is achieved by computing the node strength statistic for all nodes of the network . Node strength is defined as the sum of all weights incident to a node . In the case of a directed network , out-strength , i . e . , the sum of all outgoing link weights , and in-strength , i . e . , the sum of all incoming link weights , are calculated to provide the relative export and import risks . Node strength values can be used to rank the regions according to outgoing ( export , Fig 6B ) and incoming risk ( import , Fig 6C and 6D ) , and more critically , observe how the ranking and magnitudes of the risk change over the course of the outbreak . To determine the regions most likely to contribute to the spread of Zika virus during the epidemic , we estimated the the dynamic exportation risk using the route-level network ( Fig 6B ) . Martinique , Brazil , Colombia , Puerto Rico and Mexico are identified by the model to pose the highest risk of spreading Zika to new regions . Intuitively , the export risk is dominated by the set of counties infected earlier in the outbreak and those with high incidence rates . Martinique stands out as having the highest exportation risk , which peaks during March , corresponding to the month with the highest incidence rate ( Fig 1B ) . The highest exportation risk through 2015 is posed by Brazil and Colombia , which were the two countries reporting early and large outbreaks ( Fig 1A ) . Brazil was identified as the likely source of spread to the first few Caribbean Islands , which is consistent with the phylogenetic data [3 , 4] . Our model estimates that Brazil and Colombia became less prominent in their roles of seeding new Zika virus outbreaks from December , 2015 , to May , 2016 . This time period corresponds to the significant rise in the number of new regions reporting local Zika virus transmission ( Fig 1C ) and the rise of Zika incidence rates in many Caribbean Islands ( Fig 1B ) . The increased exportation risk posed by Martinique during this time captures this behavior ( Fig 6B ) . We similarly aggregated the route-level network at each destination to determine temporal and regional importation risk ( Fig 6C and 6D ) . The regions at highest risk of local transmission are dominated by the less developed countries and territories in Central America such as Belize , and the islands in the Caribbean , such as Haiti , Saint Lucia , Grenada , and Dominica . The high ranking of these regions is due to their low GDP ( Fig 4A ) , high human population density ( Fig 4B ) , and high vector suitability ( Fig 3 ) . For the U . S . , the states with the highest importation risks were Florida , Georgia , and South Carolina , mostly due to their high Ae . aegypti suitability ( Fig 3 ) and incoming travel volume from the affected regions [24] . Compared to much of the Americas , however , these importation risks were low , predominantly due to their high GDP ( Fig 4A ) . In fact , Florida was the only one of those three states ( along with Texas ) that reported local Zika virus transmission . These findings correspond with and reinforce previous route-level risk rankings , that is less developed regions are more likely to see local Zika virus transmission , given they meet the basic requirements for Ae . aegypti-borne virus transmission . Much of the dynamic aspects of our network model are based upon reported Zika virus cases , which are likely biased and vastly underestimated . Moreover , there is often a substantial delay ( 3-12 months ) between actual introduction of the virus and the first reported case [2–4 , 24] . The reporting inaccuracies can be attributed to the high percentage of asymptomatic Zika cases and insufficient surveillance methods , among other factors . While we cannot feasibly correct for all reporting inaccuracies , we conducted sensitivity analyses to account for delays in case reporting ( Fig 7 ) . Both three- and six-months reporting delays were considered , and the model was re-run with corresponding shifts in the data to represent each assumption . While the coefficient rankings slightly changed , likely representing better fits between local vector suitability and Zika incidence rates , the general conclusions were unchanged—low GDP was still the best predictor of local Zika virus transmission . Thus , our results appear to be robust to some reporting inaccuracies . Our work takes a major step towards improving our understanding of the risks associated with Zika virus spread and local transmission; however , there are certain limitations of this analysis that must be noted here and addressed in future research: Our work enhances our understanding of and ability to investigate the risk factors which contributed to the spread and local transmission of Zika virus during the 2015-2016 epidemic in the Americas . For each region , our model is informed by data on regional socioeconomic factors , vector habitat suitability , passenger air travel data , and epidemiological data . We constructed and implemented a dynamic extension of the GIIM to estimate the contribution of each risk factor to the likelihood of Zika virus transmission . Our model relies on a multi-agent based optimization method to estimate the parameters and a data driven stochastic-dynamic epidemic model for evaluation . The GIIM was shown to perform well based on quantitative metrics . Our results from the final model indicate the spread and local transmission of Zika virus was quite multifaceted . As expected , regional attributes influencing vectors ( Ae . aegypti suitability ) , hosts ( human population density ) , and viruses ( Zika incidence rates at origin of travel ) all contributed to the likelihood of establishing local mosquito-borne transmission . Passenger air travel volume , however , was notably less impactful that the other attributes . Therefore , rather than travel restrictions , we predict that mosquito control will be more effective at reducing Zika virus introductions leading to local transmission . This debate recently arose during the 2016 Rio Summer Olympics where some wanted to ban the games to prevent further Zika virus spread [62] . Our results suggest that additional travel for the Olympics was highly unlikely to make a significant impact . From our model , the coefficient most associated with Zika virus transmission was the regional GDP per capita , where a lower GDP corresponded to higher transmission risk . Although GDP does not directly influence transmission , it likely serves as a proxy for mosquito-host interactions [63] and surveillance activities . For example , people living in poverty often do not have the means to protect themselves from host seeking mosquitoes , such as air conditioning and screened windows common in higher income areas . Findings by Netto et al . [64] of higher Zika virus seroprevalence in areas with lower socioeconomic status further support our association . This , now evidence based conclusion that Zika and other Ae . aegypti-borne viruses should be considered diseases of poverty , is also consistent with other expert opinions [37 , 65 , 66] . The two most significant risk factors identified in our work , namely GDP and population density , are often excluded in geographic risk profiling of Aedes vector-borne diseases , and should be considered in future analysis . Our model is not specific for Zika virus and could easily be employed for other mosquito-borne viruses , such as dengue and chikungunya , with sufficient epidemiological and entomological data . Furthermore , the model could be adapted as a tool to inform real-time policy decisions regarding resource allocation for destination-based surveillance and vector control . | Since May 2015 , when Zika was first reported in Brazil , the virus has spread to over 60 countries and territories , and imported cases of Zika have been increasingly reported worldwide . However , there is still much uncertainty behind the mechanisms which dictated the rapid emergence of the epidemic . This work introduces a novel modeling framework to improve our understanding of the risk factors which contributed to the geographic spread and local transmission of Zika during the 2015-2016 epidemic in the Americas . The model is informed by data on regional socioeconomic factors , mosquito abundance , travel volumes , and epidemiological data . As expected , our results indicate that increased presence of mosquitoes , human hosts , and viruses increase the risk for mosquito-borne virus transmission . Passenger air travel , however , was less impactful , suggesting that travel restrictions will have minimal impact on controlling similar epidemics . Importantly , we found that a lower regional GDP was the best predictor of Zika virus transmission , suggesting that Zika is primarily a disease of poverty . | [
"Abstract",
"Introduction",
"Method",
"Results",
"and",
"discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"air",
"travel",
"pathology",
"and",
"laboratory",
"medicine",
"viral",
"transmission",
"and",
"infection",
"economic",
"analysis",
"pathogens",
"spatial",
"epidemiology",
"microbiology",
"social",
"sciences",
"geographical",
"locations",
"human",
"mobility",
"viral",
"vectors",
"viruses",
"rna",
"viruses",
"population",
"biology",
"infectious",
"disease",
"control",
"human",
"geography",
"infectious",
"diseases",
"geography",
"south",
"america",
"medical",
"microbiology",
"epidemiology",
"microbial",
"pathogens",
"economics",
"disease",
"vectors",
"brazil",
"people",
"and",
"places",
"population",
"metrics",
"flaviviruses",
"virology",
"viral",
"pathogens",
"earth",
"sciences",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"population",
"density",
"organisms",
"zika",
"virus"
] | 2018 | Inferring the risk factors behind the geographical spread and transmission of Zika in the Americas |
This work describes the use of entomological studies combined with in silico models ( computer simulations derived from numerical models ) to assess the efficacy of a novel device for controlled release of spatial repellents . Controlled Release Devices ( CRDs ) were tested with different concentrations of metofluthrin and tested against An . quadrimaculatus mosquitoes using arm-in cage , semi-field , and outdoor studies . Arm-in-cage trials showed an approximate mean values for mosquito knockdown of 40% and mosquito bite reduction of 80% for the optimal metofluthrin formulation for a 15-minute trial . Semi-field outdoor studies showed a mean mortality of a 50% for 24 hour trial and 75% for a 48 hour trial for optimal concentrations . Outdoors studies showed an approximate mean mortality rate of 50% for a 24 hour trial for optimal concentrations . Numerical simulations based on Computational Fluid Dynamics ( CFD ) were performed in order to obtain spatial concentration profiles for 24 hour and 48 hour periods . Experimental results were correlated with simulation results in order to obtain a functional model that linked mosquito mortality with the estimated spatial concentration for a given period of time . Such correlation provides a powerful insight in predicting the effectiveness of the CRDs as a vector-control tool . While CRDs represent an alternative to current spatial repellent delivery methods , such as coils , candles , electric repellents , and passive emanators based on impregnated strips , the presented method can be applied to any spatial vector control treatment by correlating entomological endpoints , i . e . mortality , with in-silico simulations to predict overall efficacy . The presented work therefore presents a new methodology for improving design , development and deployment of vector-control tools to reduce transmission of vector-borne diseases , including malaria and dengue .
Vector-borne diseases represent a global public health threat . More than 1 million people die annually of vector-borne diseases . Malaria alone is responsible for 400 , 000 deaths a year , and most cases are children under five years of age [1–3] . Current vector control techniques can be divided into contact repellent and spatial repellent products [4–7] . Current strategies against mosquitoes include products applied to the skin , insecticide treated nets ( ITNs ) and indoor residual spraying ( IRS ) . Some topical repellents suffer from undesired odor and texture , resulting in poor user acceptability . Topical repellents presence lowers in time due to skin washing , sweating , clothe rubbing or skin absorption , requiring several applications per day , and thus are limited in duration and rely on user compliance [8 , 9] . ITNs and IRS are inexpensive , effective indoors and when individuals remain in the confines of the treated areas ( building ) , however neither are an option for outdoors or active users outside of the residence . [10] . Spatial repellent products include electrical repellents , passive emanators based on impregnated strips , candles , and , coils [11] . Electrical repellents , including motorized fans , require power to operate , making their widespread use limited . In addition , many of these devices require frequent cartridge replacement . Passive emanators using impregnated strips are limited to a single compound showing limited efficacy only for short periods of time [12 , 13] . Candles are based on essential oils , whose efficacy has been questioned [14] . Furthermore , candles and coils require an open flame or ember to operate , representing a fire hazard , and are limited in duration , requiring frequent replacement . Coils are the most widely used products for low income countries , where electricity is not widespread . Currently , the number of coils sold worldwide is estimated at 30 billion yearly at estimated price range of US$ 0 . 025–0 . 1 per unit [15] . Their competitive cost together with their ease of use provides compelling reasons to be considered as one of the most widely accepted consumer products for vector control . Coils exhibit a number of disadvantages , including: the Active Ingredient ( AI ) represents less than 1% of the overall device volume or mass; it requires an open flame starter; it is a fire hazard during operation; it poses respiratory-related health risks , and it lacks continuous release that limits their use to 4–12 hours per unit [16 , 17] . In addition to the available armamentarium of spatial repellent devices , active devices based on Micro-Electro-Mechanical Systems ( MEMS ) , have been reported as a potential tool against mosquitoes [18] . In spite of their remarkable advantages , including low cost , batch fabrication , small form factor and the ability to actively control the delivery process , MEMS require power to operate , and they are limited in duration due to their limited payload . Therefore , there is a current need for a flameless , passive , cost effective , multi-chemistry delivery method for spatial repellency that provides sustained and safe protection over weeks in order to increase adherence of use , and ultimately further reduce risk of vector-borne diseases . Controlled Release Devices ( CRDs ) were developed to overcome some of the limitations of current spatial repellent ( SR ) delivery systems . Present CRD design uses a set of wells where formulated AIs are placed , each well is capped by a lid that forms a chamber . The chamber lids integrate a set of tightly calibrated pores to control the release of the formulated AI . The CRD also includes an exothermic process that is activated when the device is open and exposed to air via inlets to provide an increase release rate . Fig 1A and 1B show an image of a CRD and its internal components . The advantages of the CRD are: no electrical power , no battery; no open flame , making it safer than coils and candles; manufacturable with biodegradable materials , making it environmentally friendly; it can last up to two weeks; it can store multiple AIs . Present CRD design can be used in indoors or outdoors applications , and provides a cost-effective solution that can be mass produced and easily deployed . To maximize CRDs performance , high volatility is required in the AI to be used while it has to be effective against Mosquitoes . As pyrethroids compounds are known to have high efficacy in protection against mosquitoes , metofluthrin was chosen [19] . Metofluthrin is commonly referred to as a Spatial Repellent ( SR ) , but it is in fact considered an insecticide by the Environmental Protection Agency ( EPA ) , which has oversight to approve new AIs in the US . Herein we describe the systematic approach used in the CRD development , summarized in Fig 2 . The first step was the physical characterization of several metofluthrin formulations to the evaporation rate of the AI . This data is then fed into a Computational Fluid Dynamics ( CFD ) model to predict estimated concentrations of AI in the space around the device . First generation CRDs were manufactured using 3D printing Stereo Lithography Apparatus ( SLA ) technology for rapid prototyping and eventually second generation CRDs were subsequently manufactured using micro-injection molding , a cost effective technique for mass production . CRDs were tested in an arm-in-cage study , performed to determine the optimal metofluthrin concentrations in a 15 minute trial . Knockdown and bite inhibition were used as entomological endpoints . Once the optimal metofluthrin formulation was determined , CRDs were tested with such formulation for semi-field studies in order to correlate the mortality with the AI concentration in space for 24 and 48 hour periods . Mortality was selected as the preferred entomological endpoint since it can be used to correlate estimated concentration with highly specific spatial locations for a given time point . The established correlation between mortality and estimated metofluthrin concentration can be used as a tool to further tailor the CRD design and form of deployment , e . g . number and distribution of CRDs , in order to target a protective volume with a desired mortality . Lastly , outdoor studies were performed to evaluate the performance of CRDs .
Physical characterization of metofluthrin solutions was performed in order to characterize metofluthrin release rates . 50 mL falcon tubes containing metofluthrin with isopropyl alcohol ( IPA ) as a solvent and at the following concentrations: 1% , 5% , 10% , 30% , 50% ( v/v ) , were left open for evaporation rate determination at room temperature conditions . IPA at 99% v/v , was used as solvent to increase metofluthrin volatility . IPA was chosen as solvent due to metofluthrin high solubility in it ( 313 . 2 gr/L ) , high volatility and low toxicity [19] . The total mass and total volume of remaining metofluthrin solution in each sample tube for 12 , 24 , 48 and 96 hours was measured using analytical balance and calibrated volumetric measurements under laboratory controlled temperature conditions ( 20–25°C ) . Changes in relative concentrations of metofluthrin and IPA were determined via Mass Spectroscopy ( MS ) , estimating concentration of active isomers of metofluthrin E and Z . Using the same approach , the release rate of metofluthrin was determined . Water was estimated in the solution due to the hygroscopic nature of metofluthrin and IPA . Estimated metofluthrin release rates as a function of concentration were calculated from these studies . The rates were then scaled to the device geometry , calculated by multiplying by the ratio of the surface areas of the CRD to the falcon tube , in order to serve as input for the in-silico model to model spatial concentration evolution with time . CRDs preliminary versions were 3D printed , then manufactured using micro-injection molding in Zytel st801 . CRDs design consisted of an exothermic reactor bottom containing a 50 mL reservoir to store an exothermic material ( based on iron oxides ) that acted as a heat source to increase volatilization of AI after activation and hence distribute AI in the volume to be protected faster . The device also included a middle reservoir array that included seven 0 . 5 mL reservoirs to store the AI . The device also included a top layer comprising a membrane with 30 inlet pores , each 1 mm in diameter , to allow oxygen diffusion into the exothermic reactor , and 1 , 000 outlet pores , each with 200 μm diameter pores for controlled release of the AI . The exothermic reactor relied on oxygen to start the exothermic reaction , leading to a local increase in temperature of up to 55°C for eight hours and thereby enhancing the volatilization of the AI . Once cooled down , device mass rate was high enough to keep the protective volume until AI depletion . Controlled release relies on evaporation of the AI formulation through the membrane pores until the AI reservoirs are depleted . The release rate of AI is dependent on the open surface area . Metofluthrin , an effective spatial repellent , was selected due its high efficacy , low toxicity profile and relatively high vapor pressure at room temperature [20] . CRDs were vacuum sealed , and an oxygen absorber was incorporated inside the package to further reduce the presence of any oxygen to prevent the activation of the exothermic reaction . An in-silico model was developed to estimate metofluthrin spatial concentration distribution in a domain to match the observed mosquito mortality found during the entomological tests . The semi-field studies in tents configuration was chosen to reduce the air movement uncertainty that arises in an open domain . The domain consisted of an exterior volume where wind actual measured average speed and direction conditions were set . The tent was placed inside the domain with open doors to allow air to enter the interior domain . Fans present during the test were modelled inserting a speed jump at the fan location enforcing fan flow . Mass point sources were placed at the proper location with the measured device mass rate . A device is considered a point source due to its relative small size compared to tent volume . The model was developed using Computational Fluid Dynamics techniques in ANSYS [21] with k-ε turbulence for flow simulation and an implicit scalar transport scheme to tackle metofluthrin advection and diffusion in a transient solution . Changes in temperature do not affect significantly AI distribution , since natural convection dominates the mass transfer . From a fluid mechanics perspective , the estimated low AI concentrations will not lead to any buoyancy effect .
To characterize metofluthrin release rates for each of the tested formulations , as shown in Fig 3A , metofluthrin ( AI ) , IPA ( solvent ) , and water mass were integrated in the calculations to determine the fraction of each of these components as a function of time . Fig 3B shows the mass of IPA as a function of time , Fig 3C shows the mass of water as a function of time , and Fig 3D shows the mass of metofluthrin as a function of time . Water content was studied together with IPA mass and metofluthrin mass due to its hygroscopicity . IPA evaporation rate was similar for the proposed formulations , while metofluthrin increased its evaporation rate with the IPA fraction , which suggests that IPA increases the metofluthrin evaporation rate . Metofluthrin evaporation rate was calculated as the change in mass over time , shown in Fig 3E . In this plot , it is possible to observe that all curves converge towards the same rate level , which represents the 100% metofluthrin ( beyond 48 hours ) . Selection of the optimal concentration to be used in the CRDs required the Arm-in-Cage studies as the first step . Experimental setup is shown in Fig 4A and 4B . Fig 4C shows the average percentage knockdown and Fig 4D shows the average percentage of mosquito bites as a function of concentration , ranging from 1% to 100% metofluthrin . It is possible to observe that knockdown increases with metofluthrin concentration while bites are reduced , with more data dispersion in knockdown than in bites . A test with a combination of the two best performing concentrations was performed . One device with 30% metofluthrin and other one with 100% metofluthrin were tested . It was found that such combination performed even better than the best result obtained with two identical CRDs . This improvement may be attributed to the complementary of the different release kinetic profiles . One possible interpretation is that the first period of high evaporation was mainly driven by the 30% metofluthrin formulation , complemented by a second period of evaporation primarily driven by the 100% metofluthrin . Semi-field studies were carried out in tents after the optimization process performed in arm-in-cage studies . Fig 5A summarizes the experimental setup . Experimental results are shown in Fig 5B for 24 hours and in Fig 5C for 48 hours ( N = 4 ) . For each tent location a bar plot showing mosquito mortality at each of the three defined heights is shown , together with their average mortality per tent location ( yellow ) and the average control defined as a 100% IPA loaded CRD ( green ) . A uniform mortality distribution is observed though locations and heights . A minimum mortality of 50% can be observed in the first 24 hours , reaching 75% in 48 hours , with no significant differences between locations and heights . From the CFD model , simulated metofluthrin concentrations at each location and height are plotted for 24 hours and 48 hours in Fig 5D and 5E , respectively . Furthermore , the estimated spatial concentrations obtained from the CFD simulations versus mortality for every location are plotted for 24 hours and 48 hours in Fig 6A and 6B , respectively . When analyzing this correlation plot , it was found that the lower and middle level pouches seem to require less concentration to reach the same mortality as the higher level pouches . Less concentration was required to be effective in 48 hours than in 24 hours , which could be attributed to the longer exposure of the mosquitoes to the metofluthrin . Based on these plots , a linear regression analysis was performed to establish the correlation between mosquito mortality and metofluthrin concentration . The metofluthrin concentration for 100% mortality can therefore be calculated , resulting in 0 . 234 ppm for 24 hours and 0 . 097 ppm for 48 hours . Fig 6C shows the estimated iso-surface plots plotted for 24 and 48 hours . These iso-surfaces provide the spatial limit concentration found such that inside these convex surfaces a targeted mosquito mortality of 100% would be guaranteed . Finally , Fig 7A and 7B shows the experimental set up for the outdoor experiments and Fig 7C the spatial location of the pouches . The bar chart plotted in Fig 7D shows the average mosquito mortality per pouch for the distances and heights evaluated from the CRDs . Results were plotted to show average mosquito mortality per level for a given distance from the source . Controls were also shown . A slight distance effect is observed showing a decrease in mortality with almost no dependence on pouch height . Mortality dispersion in not significant across directions , even considering the unrestricted air movement .
In silico modeling provides a powerful multi-dimensional tool to estimate AI concentration as a function of targeted mortality , for a desired 3D space in a given period of time and environmental conditions , which include 3D boundary conditions , temperature , and wind velocity . A correlation was established between simulated average concentrations and mosquito mortality obtained from semi-field experiments . Additional tests to address repellency and bite inhibition in open spaces , such as Human Landed Catches , could be performed in the future . Both parameters are associated with protective surfaces what makes them more complex to correlate with a spatial distribution . The ability to correlate metofluthrin concentration with mosquito mortality in a 3D space as a function of time could potentially allow to customize SR delivery based on a target mortality and defined environmental conditions over a defined region . It is therefore possible to obtain predictive behavior of devices in terms of efficacy over a 3D space , defining a bubble of protection , over a period of time , while monitoring toxicity thresholds . Arm-In-Cage trials demonstrated that formulations of metofluthrin with IPA at 30% and 100% provided the highest percentage of knockdown in the 40–50% range and bite inhibition in the 70–90% range . Semi-field studies were performed to show the performance of CRDs in a semi-outdoor environment . High mosquito mortality rates in the range of 60–90% relative to an independent control over a period of 24 and 48 hours validated the use of CRDs for protection against mosquito bites . It was also possible to plot the mortality ( Fig 5B and 5C ) per pouch , as well as simulations results for spatial concentration per pouch ( Fig 5D and 5E ) once the release rate for each device was empirically estimated . These projected concentrations allowed to estimate the performance of device as these values were correlated with the morality values . The established correlation provides a powerful tool for projecting device performance . Outdoor experiments provided another insight in the use of CRDs as a SR protection tool . Experimental results showed mosquito mortality in pouches in the range of 40–60% relative to an independent control over a period of 24 hours for distances of up to 2 . 5 m from the devices . The presented design of CRD can accommodate volumes of up to 20 mL of AI , which represents approximately about half of the device total volume . Such a CRD version was designed to have an effective persistence ( duration ) of up to 2 weeks of continuous usage . The proposed cost of the device will be in the range of US$0 . 25–0 . 5 per unit for volumes of 1–10 million units . CRDs do not require electrical power , and do not constitute a fire hazard . Moreover , CRDs future material selections could include Mirel , a biodegradable polymer , or other selection of advanced polyhydroxyalkanoate polymers ( PHAs ) , or even paper-based versions as an environmental friendly solution . Future device designs will include a small transparent indicator with a dye to show remaining device capacity . The development and testing of a novel type of SR delivery system was introduced starting from idea conceptualization to formulation development , followed by in silico model , device design and manufacturing and entomological studies including arm-in-cage , semi-field and field experiments . CRDs were designed with a novel methodology that integrated combination of entomological endpoints and computational models to target efficacy and spatial protection as part of the design requirements . This multidisciplinary method allowed for a quantitative approach to device development and optimized performance . The design method allows for tailoring release kinetic profiles and field distributions for public health interventions in buildings , facilities and open areas . Spatial coverage can be achieved by the number and distribution of devices deployed . The duration can be tailored by the device capacity . Experimental results have shown the potential use of CRDs as the next generation SR device . CRDs represent a simple and cost-effective solution for enhanced protection against vector-borne diseases . | Spatial Repellents ( SRs ) represent another tool to fight vector-borne diseases , such as malaria and dengue . Newly developed active ingredients were designed to repel or kill vectors in space , creating a shield effect , unlike topical repellents , such as DEET , that rely on vectors to be near or in physical contact with protected target . Metofluthrin and transfluthrin are examples of active ingredients ( AIs ) designed as SRs against mosquitoes . The efficacy of SRs heavily depends on the delivery method . Currently , there is a lack of fundamental understanding of effectiveness of SR delivery methods . Current delivery modalities of SR do not rely on quantitative models to estimate targeted efficacy , making the end-user overshoot or undershoot the dosage required for protection . Optimizing the dosage over time is critical to obtain protection for a given space in a giving period of time , as well to prevent AI resistance in the long run . The key is therefore to deliver just enough dosage needed to repel or kill the vectors . The presented work provides a novel approach to predict performance of SRs based on an experimental-computational methodology to quantify effectiveness of controlled release devices as a function of AI physical properties ( e . g . volatility ) , device design parameters combined with physical variables , and environmental conditions ( e . g . temperature , wind velocity ) . This work therefore provides the groundwork for estimating quantitative effectiveness of SR delivery methods against mosquitoes . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"death",
"rates",
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"classical",
"mechanics",
"fluid",
"mechanics",
"condensed",
"matter",
"physics",
"vector-borne",
"diseases",
"animals",
"simulation",
"and",
"modeling",
"vaporization",
"population",
"biology",
"insect",
"vectors",
"zoology",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"agrochemicals",
"fluid",
"dynamics",
"phase",
"transitions",
"disease",
"vectors",
"insects",
"continuum",
"mechanics",
"agriculture",
"arthropoda",
"physics",
"insecticides",
"population",
"metrics",
"mosquitoes",
"eukaryota",
"entomology",
"biology",
"and",
"life",
"sciences",
"evaporation",
"species",
"interactions",
"physical",
"sciences",
"organisms"
] | 2019 | A combined experimental-computational approach for spatial protection efficacy assessment of controlled release devices against mosquitoes (Anopheles) |
Unlike the pandemic form of HIV-1 ( group M ) , group O viruses are endemic in west central Africa , especially in Cameroon . However , little is known about group O’s genetic evolution , and why this highly divergent lineage has not become pandemic . Using a unique and large set of group O sequences from samples collected from 1987 to 2012 , we find that this lineage has evolved in successive slow and fast phases of diversification , with a most recent common ancestor estimated to have existed around 1930 ( 1914–1944 ) . The most rapid periods of diversification occurred in the 1950s and in the 1980s , and could be linked to favourable epidemiological contexts in Cameroon . Group O genetic diversity reflects this two-phase evolution , with two distinct populations potentially having different viral properties . The currently predominant viral population emerged in the 1980s , from an ancient population which had first developed in the 1950s , and is characterized by higher growth and evolutionary rates , and the natural presence of the Y181C resistance mutation , thought to confer a phenotypic advantage . Our findings show that although this evolutionary pattern is specific to HIV-1 group O , it paralleled the early spread of HIV-1 group M in the Democratic Republic of Congo . Both viral lineages are likely to have benefited from similar epidemiological contexts . The relative role of virological and social factors in the distinct epidemic histories of HIV-1 group O and M needs to be reassessed .
Human Immunodeficiency Virus Type 1 ( HIV-1 ) is comprised of four groups ( M to P ) , each originating from a distinct cross species transmission event from Simian Immunodeficiency Virus ( SIV ) variants circulating in apes [1 , 2 , 3] . The major group ( M ) has spread worldwide from Central Africa during the second part of the 20th century [4 , 5] , while groups N and P are extremely rare . These latter two groups have arisen more recently and have only been identified so far in 15 and 2 patients respectively [1] , all but one from Cameroon . Finally , despite group O’s origin being estimated to be about the same time period as group M [6] , the group O ( outlier ) epidemic is mostly restricted to Cameroon , and has remained stable since the 1990s , whereas group M has been spreading dramatically [7 , 8 , 9] . Little is known about natural history of group O infection , but the limited follow-up data available [7 , 10] indicate that , as with group M , horizontal as well as vertical transmission contributes to its spread , and untreated patients show high plasma viral load , leading to a loss of CD4 T cells and eventual progression to AIDS . Thus , the natural history of group O infection seems to be similar to that of group M , even though some studies have shown distinct virological properties such as a lower replication capacity [11 , 12] or failure to counteract some cellular restriction factors [13 , 14] . There is high genetic distance between M and O strains , with 67% , 73% and 56% sequence identity between group M HXB2 [15] and O ANT70 [16] prototype strains in gag , pol , and env genes , respectively . As a result , diagnosis and follow-up of group O infections require adapted tools [1] . Group O natural polymorphism also has an impact on treatment options , since most strains naturally present the Y181C mutation in the Reverse Transcriptase ( RT ) conferring resistance to Efavirenz and Nevirapine ( first generation Non Nucleoside RT Inhibitors , NNRTIs ) . Of particular note , these molecules are part of the most common first line antiretroviral therapy combinations used in Cameroon . Taken together , these group O characteristics can lead to delayed diagnosis , underestimated viral loads or treatment failure , if the nature of the group O infection status is not taken into account . Not only is group O highly distant from group M , but over nearly a century since its introduction into the human population , a high level of intra-group genetic diversity has developed and several attempts have been made to characterize it . Different classifications have been proposed , and these have defined 3 clades [17 , 18] , 5 clusters [19] and more recently , only 2 lineages ( C181 or Y181 ) based on the residue at the RT position 181 [20] . The relevance of these different classifications , which were based on few sequences and never compared to each other , needs to be evaluated . More importantly , the processes that led to this apparent complexity are still to be understood . No significant group O epidemic has been described outside Cameroon , and only sporadic cases appear in other African countries ( mostly west central Africa ) , in the US and in Europe [1] . In France , the RES-O survey network has identified 143 patients infected with group O since 1992 [10 , 21] , the largest series of group O infections identified outside of Cameroon . Most of these patients originate or are linked to patients originating from Cameroon , due to historical links between France and this region . This absence of epidemic spread outside of the Cameroon is surprising , since group O infections have been identified on three different continents , as far back as 20 years ago in North America and 50 years ago in Western Europe , and an estimated presence close to a century old in Africa . Interestingly , it has recently been proposed that group M expansion in the capital of the Democratic Republic of Congo ( DRC ) Kinshasa had benefited from different contextual changes in the 1960s , leading to a dramatic increase of the number of infections at that time [5] . The authors did not observe such a change in group O growth rate , thus hypothesising that this explained the different epidemiological histories of groups M and O . Here we have used the largest set of HIV-1 group O sequences assembled to date , obtained from 190 patients sampled in France or in Cameroon on a time scale of 26 years , to better understand group O emergence and evolution , by investigating the dynamics of their diversification and its consequences .
The maximum likelihood tree obtained from 190 concatenated sequences showed that a large proportion of the strains fall in a major subgroup . The short branch lengths in this major subgroup when compared to the long ones in the minor subgroup ( S1 Text and S1 Fig ) give the tree a comet-like shape ( Fig 1A ) , as opposed to the well-defined double star structure of the pandemic group M subtypes ( Fig 1B ) . Due to this particular topology , we defined the major subgroup ( N = 147 ) as corresponding to the "comet head" or H strains , and the minor subgroup ( N = 43 ) , as corresponding to the "comet tail" or T strains . Subclusters could be observed among the H strains ( H1 , H2 and H3 ) and the T strains ( T1 and T2 ) . This classification encompasses the previous ones that were partially discordant ( see S2 Text and S2A and S2B Fig ) . We used 154 sequences for which all three gene fragments were obtained from a single sample of known sampling time to investigate group O origins and dynamics over time . Different coalescent population growth models ( constant size , exponential growth , logistic growth and Bayesian skyline ) gave consistent estimates for group O’s time to most recent common ancestor ( tMRCA ) of around 1930 ( Fig 2 , black curves ) , with 95% highest posterior densities ( 95% HPD ) ranging from 1914 to 1944 . Interestingly , Bayesian Skyline Plots showed that group O genetic diversity had gone through an alternation of slow and fast growth phases ( Fig 3A ) . Two waves of exponential growth were observed , the first around 1950 and the second , longer and more important , starting in the late 1970's and ending in the early 1990s . While the first wave could be observed when investigating all of the 154 sequences ( Fig 3A ) , it did not appear when only including the H strains and the two minor subclusters observed among the T strains , T1 and T2 ( Fig 3B ) . These results indicate that the first wave represents the development of an ancestral level of genetic diversity , and the second wave the emergence of subpopulations such as H strains . We then investigated the date of subgroup H MRCA using the same 4 growth models , which consistently gave an MRCA estimated around 1945 ( best fitting model 95% HPD: 1933–1955 ) ( Fig 2 , green curves ) . This suggests that H strains MRCA was present among the background of genetic diversity that had arisen during the first exponential growth phase in the 1950s . Among these H strains , we observed some important and strongly supported subclusters H1 , H2 , H3 , which all together represented 34% of the group O strains studied herein . Their respective MRCA estimates were even more recent , all being estimated to have appeared between 1975 and 1980 ( Fig 2 , blue , red and yellow curves ) , the very beginning of the second exponential growth phase . When studying the distribution of the Y181C resistance mutation among the 100 samples collected from NNRTI-naïve patients ( 77 H strains , 23 T strains ) , we observed that this profile was naturally present in 65 strains ( 65% ) , 62 of which were H strains and only 3 T strains . Thus , 80 . 5% of the H strains naturally presented the resistance mutation and only 13% of the T strains , indicating a strong association between subgroup H and C at position 181 ( Chi-squared test: p<10E-5 ) . Moreover , among the 15 H strains presenting a 181Y residue , 11 ( 73 . 3% ) belonged to a single subcluster ( H3 ) . Bayesian analyses indicated that the mean evolutionary rate of H strains was significantly higher than that of the T strains ( Student T test: p<10E-5; see Table 1 ) . For both subgroups , the dN/dS ratios indicated a globally negative selection pressure on the genome regions investigated , with this ratio was also significantly higher in subgroup H than in subgroup T ( Likelihood Ratio Test: p = 7 . 10E-4 , see Table 1 ) . Thus , H strains evolved at a higher rate and had lower negative selection pressure than T strains . Evolutionary analyses also indicated that the mean growth rate of subgroup H was significantly higher than that of the T ( Student T test: p<10E-5; see Table 1 ) , consistent with the apparent predominance of the H strains .
Even though both HIV-1 groups O and M’s MRCAs have been estimated in the early 20th century , group O , unlike group M , has not spread globally and remains endemic in Cameroon . The causes of this are not fully understood: unlike HIV-1 groups N and P , group O has been successful enough to infect tens to hundreds of thousands people [3 , 18] and no evidence of lower transmissibility has been shown in vivo , as it has been for HIV-2 [27] . Thus , aside from potential intrinsic properties [11 , 12] , epidemiological and related contextual factors could also have played a role in this difference between the two main HIV-1 lineages [5] . Our data confirmed that group O diversification represents a continuum of diversity consistent with local diversification in a geographically restricted area ( see S2 and S3 Texts and S2 Fig ) , while the group M subtypes are mostly the result of founder effects following the introduction of single strains to dispersed geographic locations [28] . We also confirmed the existence of a predominant population ( H strains—from the comet Head ) , which was strongly associated to the C mutation at the Reverse Transcriptase position 181 , but our data demonstrated that this mutation alone is not an adequate marker for classification as proposed by Tebit et al . [20] . Indeed , it was not possible to just split the tree in two C181 and Y181 subpopulations , even if hypothesizing unlikely transmissions of strains with acquired mutations for the few T strains harbouring this Y181C mutation . The existence of this major subgroup raised questions on the evolutionary processes that could have shaped the particular tree topology . Our estimation of group O MRCA around 1930 ( 95% HPD: 1914–1944 ) is close to the previous estimates of 1920 [6] which was based on fewer sequences . This result confirms that group O MRCA was contemporaneous to that of group M , estimated in the beginning of the 20th century [4] . But unlike recent conclusions of a constant low growth rate for group O by Faria et al . [5] , we show two phases of exponential growth , the first during the 1940–1960 time window and the second one during 1970–1990 . The causes for these two phases are probably complex and multifactorial . However , the increase in growth rates around 1950 coincides with a period of high-rate of transmission of HCV in Cameroon , as shown in both epidemiological [24] and phylogenetic studies [23] , ( Fig 3C ) . In Southern Cameroon , HCV infection is a reliable marker of iatrogenic transmission , associated in elderly people with a history of exposure to medical campaigns and interventions involving unsterile procedures [29] . The 1940–1960 period saw a rise in injection practices globally [30] . Colonial medical activity in Cameroon peaked during this period , and was dismantled after independence ( 1960 ) for political and epidemiological reasons ( low incidence of sleeping sickness and yaws ) [25] . Since the first plateau identified in this study matches the decline of HCV transmission in Cameroon after 1960 , the first phase of group O rapid growth can thus reflect a scenario of iatrogenic amplification following an event of cross-species transmission , as proposed for HIV-1 M in the Congo basin region [31 , 32] . The booming city of Douala may have been a favorable epidemic context—combining iatrogenic and socio-sexual factors—for the initial diversification of group O . Indeed , this is most probably where a visiting Norwegian sailor—the first reported case of group O infection—became infected in 1962 [33 , 34] , suggesting that the virus was established in this city by the early 1960s . Two decades later , the greater and longer second phase of exponential growth started in the late 1970's and reached a plateau in the early 1990s . Though group O might have found a favourable social context for transmission during this period , urban growth per se cannot explain the exponential increase in new infections , since rates of urban growth in Yaoundé and Douala were maximal in the post-second war years ( at about 10% ) and have declined since , down to 5–7% in the 2000’s [35 , 36 , 37 , 38] . Further investigations are thus needed to understand this second phase of diversification . Intrinsic viral properties could have been involved , since the second wave of diversification is linked to the development of several viral subpopulations , some minor in the T subgroup ( T1 and T2 ) , and the H strains which became predominant . Of note , subclusters H1 , H2 and H3 , which MRCA was estimated to be no older than 1975 , now represent 34% of the strains included in this study . They have been sampled either in Cameroon or in France , at time points ranging from 1994 to 2012 , showing that they are not associated to a particular sampling time or location ( see S2 Text and S2C and S2D Fig ) . Selection pressure analysis also showed that H strains were submitted to lower negative selection pressure than T strains and evolved faster . It has also been proposed that the 181C residue in the RT could represent a fitness advantage [20] , even though this has been demonstrated in vitro only using mutants viruses from a H strain backbone . The reasons why H strains became predominant could thus be linked to favourable biological properties of these strains and/or different opportunities for diffusion . The two phases may also reflect a two-step geographic expansion of the virus in Cameroon , but the absence of a geolocalized dataset does not allow us to investigate this hypothesis further . Our data thus show that group O genetic diversity and phylogenetic topology are due to their evolution in alternating slow and fast phases that could be related to specific events in the history of Cameroon , contrasting with recent conclusions [5] . However , these variants failed to become pandemic despite two exponential growth phases . Interestingly , the end of the last exponential phase coincides with the time of introduction and spread of CRF02_AG , the predominant group M form in Cameroon , after it originated in the 1970's in the DRC [22] ( Fig 3C ) . This led to the previous observation in Cameroon of a drop in group O infections among HIV-1 positive samples in the early 1990s , when the M group rose exponentially while group O remained stable or even slightly decreased [8 , 9] . How group M became predominant over the group O epidemic in Cameroon could be due to different virological properties [11 , 12] and/or a competition between the two epidemics , as suggested for HIV-2 in West Africa [39] . This would also explain the absence of French group O epidemic ( see our phylogeographic analyses in S2 Text and S2C Fig ) , contrasting with HIV-1 group M molecular epidemiology . Most of the 143 group O patients identified originate from Cameroon , while after the long predominance of group M subtype B in France , non-B strains have been imported from sub-Saharan Africa and now circulate in patients from various risk groups [40] . Thus , the absence of diffusion of group O in France has to be explained by other factors than a lack of boundaries between different epidemiological populations . In summary , our results on group O genetic diversity support the conclusion that it cannot be divided into subtypes similar to group M , but a major subgroup ( H ) emerged from a genetically diverse minor subgroup ( T ) . Contrary to Faria's findings [5] , these two populations reflect the fact that group O underwent two waves of rapid diversification in the 20th century . Although this evolutionary pattern is specific to group O , the HIV-1 group O epidemic in Cameroon paralleled the spread of HIV-1 group M in the DRC . This is certainly linked to similar factors as those described for group M , such as iatrogenic amplification and the favorable urban context of fast-growing and cosmopolitan cities . In this light , the emergence of group M in Kinshasa may be seen as unexceptional , however expansion of groups O and M occurred in distinct epidemiological and socio-demographic contexts in Cameroon and the DRC respectively , which could have led to their different epidemiological patterns [2] . While our study reveals important information about HIV-1 group O’s emergence and evolution , investigations are still needed to understand the other reasons for their unsuccessful spread compared to that of group M—as well as for group O T strains compared to that of H strains—especially by exploring the biological properties of these divergent viruses .
Samples from 190 patients were included in the analysis , 102 of which were sampled in France , 87 in Cameroon , and 1 in Gabon . The time of sampling spanned from June 1987 to February 2012 , but was undetermined for 4 samples . In France , the samples were collected from hospitals located all across the country by the RES-O survey of the French HIV National Reference Centre in Rouen . In Cameroon , samples were collected at the Centre Pasteur du Cameroun from different parts of the country: Centre ( N = 25 ) , Littoral ( N = 7 ) , North ( N = 4 ) , and South ( N = 1 ) regions . For 50 samples , the collection site in Cameroon was not determined . The sample from Gabon was collected in Libreville . The nature of the samples analyzed was plasma or serum ( N = 173 ) , PBMCs ( N = 6 ) or supernatant from viral culture ( N = 11 ) . The viral sequences we analyzed were produced from leftover samples that had previously been collected for routine diagnosis or follow up of the patients . Thus , no additional sample was performed specifically for this study , and we did not use any information about the patients from whom this samples had been obtained . As a consequence , no consent from the patients nor approvement from ethics committee was requested . Three fragments from two genes ( prRT: pol protease and partial RT , 987 bp; int: pol partial integrase , 603 bp , gp41: env partial gp41 , 522 bp ) were amplified by nested PCR or RT-PCR , depending on the sample type , and sequenced as previously described [41] . Accession Numbers: KT198045—KT198614 . | HIV-1 group O is one of the causal agents of AIDS , together with HIV-1 groups M ( responsible for the pandemic ) , N and P ( 15 and 2 cases detected respectively , from Cameroon ) and HIV-2 groups A to I ( mostly found in West Africa ) , each group resulting from a distinct cross species transmission event from non-human primates . Even though mostly restricted to Cameroon , group O infections have been found in other African countries as well as in Europe and in the US . Due to their genetic distance from the pandemic HIV-1 group M , group O viruses still impact diagnosis , virological and therapeutic monitoring . Moreover , very few data are available on the natural history and epidemiology of these infections , as well as their genetic diversity and evolution . In particular , there is currently no explanation of the lack of spread of these variants , compared to the pandemic viruses from group M . Analysis of HIV-1 group O molecular evolution , from sequences spanning more than 2 decades , is an opportunity to better understand the phylodynamics of group O infection . We investigate it further by producing the largest set of group O sequences described . We show that the previous classifications proposed do not agree with each other and do not fit with the extensive genetic diversity of this group . We also estimate that group O MRCA existed in the 1930s ( 95% Higher Posterior Density: 1914–1944 ) , and show that group O has diversified during two successive phases that could be linked to the specific historical context of Cameroon . These results contribute to a better understanding of the factors influencing HIV evolution , especially in the local context of west central Africa and lead to new hypotheses on the limited diffusion of such variants . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | The Two-Phase Emergence of Non Pandemic HIV-1 Group O in Cameroon |
Human cytomegalovirus ( HCMV ) infection is associated epidemiologically with poor outcome of renal allografts due to mechanisms which remain largely undefined . Transforming growth factor-β1 ( TGF-β1 ) , a potent fibrogenic cytokine , is more abundant in rejecting renal allografts that are infected with either HCMV or rat CMV as compared to uninfected , rejecting grafts . TGF-β1 induces renal fibrosis via epithelial-to-mesenchymal transition ( EMT ) of renal epithelial cells , a process by which epithelial cells acquire mesenchymal characteristics and a migratory phenotype , and secrete molecules associated with extracellular matrix deposition and remodeling . We report that human renal tubular epithelial cells infected in vitro with HCMV and exposed to TGF-β1 underwent morphologic and transcriptional changes of EMT , similar to uninfected cells . HCMV infected cells after EMT also activated extracellular latent TGF-β1 via induction of MMP-2 . Renal epithelial cells transiently transfected with only the HCMV IE1 or IE2 open reading frames and stimulated to undergo EMT also induced TGF-β1 activation associated with MMP-2 production , suggesting a role for these viral gene products in MMP-2 production . Consistent with the function of these immediate early gene products , the antiviral agents ganciclovir and foscarnet did not inhibit TGF-β1 production after EMT by HCMV infected cells . These results indicate that HCMV infected renal tubular epithelial cells can undergo EMT after exposure to TGF-β1 , similar to uninfected renal epithelial cells , but that HCMV infection by inducing active TGF-β1 may potentiate renal fibrosis . Our findings provide in vitro evidence for a pathogenic mechanism that could explain the clinical association between HCMV infection , TGF-β1 , and adverse renal allograft outcome .
Human cytomegalovirus ( HCMV ) has been associated with poor renal allograft outcome in numerous seroepidemiologic studies [1] , [2] , [3] , [4] . Evidence of active CMV infection ( DNAemia , antigenemia ) or CMV disease in renal transplant recipients is also associated with poor graft outcome [5] . In a rat renal allograft model , infection with rat CMV accelerates and intensifies rejection in infected allografts as compared to uninfected allografts [6] , [7] , [8] . These studies support an association between HCMV and adverse renal allograft outcome , but the mechanisms by which HCMV contributes to renal allograft loss remain cryptic . The fibrogenic cytokine transforming growth factor-β1 ( TGF-β1 ) is present in biopsy specimens of human renal allografts undergoing rejection [9] , [10] , [11] . TGF-β1 is produced by infiltrating leukocytes during rejection and may also be produced by renal tubular epithelium [12] , [13] , [14] . TGF-β1 is expressed at higher levels in HCMV infected renal allografts compared to uninfected allografts [15] . In a rat renal transplantation model , allografts from rat CMV infected animals also contain greater quantities of TGF-β1 as compared to uninfected allografts [6] , [16] , [17] . TGF-β1 contributes to renal fibrosis in numerous animal models and in human fibrotic renal disease , by inducing epithelial-to-mesenchymal transition ( EMT ) of renal tubular epithelial cells [18] , [19] , [20] , [21] . During EMT , renal tubular cells demonstrate loss of epithelial characteristics and cellular adhesions , develop changes in the actin cytoskeleton , induce expression of fibrogenic molecules , and acquire a migratory phenotype [22] . These fibroblastoid renal tubular cells are key contributors to renal fibrosis , as inhibition of TGF-β1 mediated EMT prevents and reverses experimentally induced renal fibrosis in animal models [22] , [23] , [24] . The association between CMV and TGF-β1 in renal allografts raises the possibility that CMV might accelerate renal allograft loss via viral induction of TGF-β1 with resultant fibrosis within the allograft . Studies performed in vitro have shown that CMV induces secretion of TGF-β1 from infected fibroblasts , astrocytes , and osteosarcoma cells [25] , [26] , [27] . TGF-β1 production can also be induced by transient transfection of expression plasmids containing the HCMV immediate early 1 and 2 ( IE1 , IE2 ) genes into fibroblasts and astrocytoma cells [25] , [28] . In those studies , increases in TGF-β1 were associated with induction of TGF-β1 mRNA . However , the local effects of TGF-β1 are often controlled in vivo by activation of the extracellular latent form [29] . Known activators of latent TGF-β1 include proteases ( plasmin ) , matrix metalloproteases ( MMPs ) , thrombospondin-1 ( TSP-1 ) , and the αvβ6 and αvβ8 integrins [30] . In the HCMV infected placenta , HCMV infected endothelial cells have been shown to induce production of TGF-β1 and collagen IV via induction of αvβ6 integrin [31] . Thus , precedent exists for the possibility that HCMV infected renal cells might induce TGF-β1 production or activation in pathological settings . HCMV can infect renal tubular epithelial cells . HCMV antigens and DNA are found in renal epithelial cells in kidneys of trauma victims examined during autopsy as well as in biopsies of renal allografts , indicating that these cells can harbor HCMV in both healthy persons and allograft recipients [32] , [33] . HCMV antigens have also been detected in tubular cells of biopsies from HCMV seropositive patients with rejection [34] . Furthermore , HCMV has been detected more often in renal tubular epithelium of allograft biopsies with rejection compared to those without rejection using both immunohistochemistry and in situ hybridization [35] , [36] . Based on the epidemiologic data associating HCMV infection with long-term allograft loss , histologic evidence that TGF-β1 production is increased in HCMV infected renal allografts , and in vitro data supporting HCMV induction of TGF-β1 production by fibroblasts and other cells , we hypothesized that HCMV infected renal tubular epithelium undergoing EMT might develop a fibroblast-like phenotype with secretion of TGF-β1 . In the following studies , we demonstrate that HCMV infected renal tubular epithelial cells undergo EMT and thereafter induce active TGF-β1 production . In this system , MMP-2 mediates the extracellular activation of TGF-β1 in HCMV infected cells , and can be recapitulated by transient transfection of the HCMV IE1 or IE2 open reading frames into renal epithelial cells stimulated to undergo EMT . These data provide supportive evidence that HCMV infected renal epithelium may contribute to long-term renal allograft loss by exacerbating fibrosis via TGF-β1 , and provides a potential in vitro mechanism for the association of HCMV infection with greater TGF-β1 production in renal allografts and long-term allograft loss .
Primary human foreskin fibroblasts and the immortalized human proximal renal tubular epithelial cell line , HK-2 , were infected in parallel with HCMV strain TR at an MOI of 1 . Some HK-2 cultures were treated with recombinant active ( ra ) TGF-β1 at 15 ng/ml , starting 1 hour post-infection . Culture media and cells were harvested separately on days 1-6 post-infection and viral titers determined by the detection of early antigen fluorescent foci ( DEAFF ) assay . Separate cell cultures were infected and harvested at day 5 post-infection , and cellular lysates subjected to western blotting using a high titered human CMV immune globulin ( Cytogam ) . Fibroblasts supported productive infection with logarithmic increases in viral progeny observed in both media and cells ( Figure 1A ) . HK-2 cells also supported productive infection ( Figure 1B ) but the kinetic was linear and the virions remained cell-associated , with very few virions detectable in media . By day 3 post-infection , virus was not detectable in the media . No difference in viral growth kinetics was observed in the HK-2 cultures treated with TGF-β1 ( Figure 1C ) , compared to HK-2 cultures without TGF-β1 . Viral titers in HK-2 cells and media were similar at each time point post infection , in the presence and absence of TGF-β1 . This result differs from TGF-β1 effects upon HFFs , where HCMV replication is induced by exposure to TGF-β1 [37] . Western blotting of the infected cell lysates ( Figure 1 , insets ) indicates a similar pattern of infected cell viral proteins in both HFFs and HK-2 cells , consistent with productive infection in both cell types . Viral replication studies were repeated in primary renal epithelial cells and results similar to those in HK-2 cells were observed , with the majority of the replicating virus remaining cell-associated ( Figure S1 ) . We next investigated whether HCMV infected renal tubular cells could undergo EMT similar to uninfected cells , or whether infection caused cells to remain epithelioid in the presence of TGF-β1 . HK-2 cells were infected with HCMV strain TR , followed by incubation with raTGF-β1 for 48 hours as an inducer of EMT . Cells were fixed , permeabilized , and stained using an anti-HCMV IE1 antibody ( p63-27 ) and species-specific AlexaFluor 488 conjugated secondary antibody , and the AlexaFluor 594 conjugated phalloidin ( Figure 2A ) . Imaging by confocal microscopy demonstrated that HCMV infected renal tubular epithelial cells exhibited cuboidal structural actin cytoskeleton , similar to uninfected cells ( Figure 2A , left column ) . The HCMV infected renal epithelial cells appeared morphologically indistinguishable from uninfected cells and could only be distinguished by immunostaining for HCMV IE1 . After raTGF-β1 stimulation , HCMV infected epithelial cells showed changes in actin cytoskeleton , developing an elongated mesenchymal phenotype associated with parallel actin stress fiber formation and loss of cuboidal epithelial morphology . The raTGF-β1 induced cytoskeletal changes appeared similar in both HCMV infected and uninfected renal tubular cells ( Figure 2A , right column ) . Imaging results were similar using primary renal tubular epithelial cells ( data not shown [dns] ) . These results showed that HCMV infected renal tubular cells can undergo morphologic changes consistent with EMT , similar to those described for uninfected cells [22] . In the next series of experiments , primary renal tubular cells were left untreated , or were infected with HCMV strain TR , followed by incubation with or without raTGF-β1 for 48 hours . This permitted analysis of effects of HCMV alone , TGF-β1 alone , or the combination of HCMV and TGF-β1 upon renal tubular cells . Cells were lysed and equivalent protein lysates were assayed for expression of HCMV IE1 , E-cadherin , vimentin , or actin by immuno-blotting ( Figure 2B ) . At baseline , both HCMV uninfected and infected cells expressed E-cadherin but not vimentin , consistent with an epithelial phenotype . Following raTGF-β1 stimulation , E-cadherin was no longer present , and vimentin was induced by both HCMV uninfected and infected cells , consistent with a mesenchymal phenotype . Immunoblotting for phosphorylated SMAD2 ( p-SMAD2 ) in nuclear extracts revealed the presence of p-SMAD2 in nuclei of raTGF-β1 treated cells , both in the presence and absence of HCMV infection , whereas cells that were not treated with raTGF-β1 did not contain nuclear p-SMAD2 . Again , this result was consistent with the induction of TGF-β1 dependent SMAD signaling in HCMV infected cells . HCMV infected cells expressed IE1 as expected , and actin staining showed equivalent protein loading for all samples . Taken together , these results were consistent with findings from imaging studies ( Figure 2A ) and indicated that HCMV infected renal tubular epithelial cells could undergo loss of E-cadherin expression , acquisition of vimentin expression , and SMAD2 phosphorylation in the presence of raTGF-β1 . The phenotypic changes were consistent with changes indicative of EMT in uninfected renal tubular epithelium [22] , [38] . To characterize the phenotype of HCMV infected HK-2 cells before and after EMT , HK-2 cells were examined at baseline , or after HCMV infection , in the presence or absence of raTGF-β1 . Thus , effects of HCMV alone , TGF-β1 alone , or HCMV and TGF-β1 together could be compared to HK-2 cells at baseline . RNA was extracted from cell lysates , and RT-PCR performed using a commercial PCR array to detect transcripts associated with extracellular matrix molecules . Transcripts from cells under various experimental conditions were compared to those produced at baseline by uninfected , unstimulated HK-2 cells . Compared to uninfected cells , cells infected with HCMV showed less than 10-fold induction of many mRNA transcripts encoding fibrogenic matrix proteins represented in this array ( Figure 2C , grey bars ) . This is consistent with the light microscopic appearance of HK-2 cells infected with HCMV , which appeared epithelioid and morphologically indistinguishable from adjacent uninfected HK-2 cells ( Figure 2A ) , and confirmed that , in the absence of raTGF-β1 stimulation , HCMV infected HK-2 cells did not induce global transcriptional changes suggestive of EMT . In contrast , after exposure to raTGF-β1 , transcriptional up-regulation of a number of fibrogenic molecules was observed , indicating induction of EMT in both HCMV uninfected and infected cells ( Figure 2C ) . Some transcripts , such as for fibronectin , MMP-9 , and the αv integrin , were highly upregulated in both uninfected ( hatched bars ) and HCMV infected ( black bars ) cells stimulated with raTGF-β1 . A few transcripts , such as for TIMP-2 , were induced to higher levels in HCMV uninfected cells compared to HCMV infected cells after raTGF-β1 stimulation . Transcripts encoding many fibrogenic molecules ( ADAMTS1 , TGF-β1 , β-catenin , collagens , MMPs , TIMP-1 , thrombospondins ) were induced to greater levels in HCMV infected cells as compared to uninfected cells after EMT . These results suggested that HCMV infected renal tubular cells expressed mRNA transcripts of numerous fibrogenic proteins at similar or higher levels than uninfected renal tubular cells after induction of EMT with raTGF-β1 . These results indicated that HCMV infected cells were capable of exhibiting the fibrogenic phenotype of EMT , and that HCMV infection does not reduce or prevent the EMT phenotype in these cells . To confirm the results from those studies using PCR array , individual RT-PCR assays were repeated in separate experiments using both HK-2 cells and primary renal tubular epithelial cells , utilizing commercial primer-probe sets for several of the mRNAs ( Figure 2C , asterisks ) that were upregulated in HCMV infected HK-2 cells after EMT ( fibronectin , MMP-9 , ADAMTS1 , TGF-β1 , collagen 5A1 , MMP-2 , and thrombospondin-1 ) . These individual PCR assays validated the upregulation of these mRNA transcripts observed in the PCR array ( Figure S2 ) . The mRNAs for MMP-9 and ADAMTS1 showed a lesser log induction in the individual assays compared to the PCR array ( 103–104 fold induction vs . 106–07 fold induction ) ; however , the individual assays did confirm that these mRNA transcripts were highly upregulated after TGF-β1 stimulation , consistent with the array results , in both HK-2 cells and primary PTECs ( Figure S2 ) . Taken together , these results demonstrate by morphologic and phenotypic assays that HCMV infected cells can undergo EMT similar to uninfected cells . HCMV infection does not prevent or diminish the EMT phenotype as compared to uninfected cells . Because of the association of TGF-β1 with HCMV infection in renal allografts , we next explored whether HCMV infected HK-2 cells could produce TGF-β1 . HK-2 cells were untreated , or infected with HCMV strain TR at an MOI of 1 and/or stimulated with raTGF-β1 for 48 hours to induce EMT . Cells were washed extensively to remove exogenous TGF-β1 , fresh media applied and samples harvested at 24 hours after washing . Samples were assayed for active or total ( active + latent ) TGF-β1 activity using a published luciferase reporter bioassay ( Figure 3A , B ) and a commercial human TGF-β1 ELISA ( Figure 3E ) . Because all exogenous raTGF-β1 was washed from the cultures and only fresh media not containing exogenous TGF-β1 was assayed , the TGF-β1 observed in samples was presumed to be derived from the cultured cells . No TGF-β1 was detectable in the serum-free media used for HK-2 cell culture by luciferase bioassay or ELISA ( dns ) . At baseline , non-infected HK-2 cells produced some detectable active TGF-β1 ( HCMV TR-/raTGF-β1- ) ( Figure 3A ) . HCMV infected HK-2 cells ( HCMV TR+/raTGF-β1- ) also produced similar amounts of active TGF-β1 compared to uninfected cells , indicating that HCMV infection alone did not induce de novo TGF-β1 production in these epithelial cells ( Figure 3A ) . This result differed from results shown by others for HCMV infected fibroblasts and astrocytes , where HCMV infection induced TGF-β1 production , and may reflect biological differences between infection of those cell types as compared to renal epithelial cells and possibly the strain of virus utilized in these previous studies [25] , [26] . Uninfected HK-2 cells stimulated with raTGF-β1 ( HCMV TR-/raTGF-β1+ ) also produced similar levels of active TGF-β1 compared to both infected and uninfected cells not exposed to raTGF-β1 , indicating that EMT alone did not induce HK-2 cells to produce additional amounts of active TGF-β1 ( Figure 3A ) . This result also confirmed that the exogenous raTGF-β1 , used to induce EMT , did not carry over in detectable quantities to our assay for active TGF-β1 . In contrast , HCMV infected HK-2 cells stimulated with raTGF-β1 ( HCMV TR+/raTGF-β1+ ) produced significantly more active TGF-β1 ( p≤0 . 01 ) than did cells in other conditions ( Figure 3A ) . The quantity of total TGF-β1 was similar in all conditions ( Figure 3B ) , indicating that changes in active TGF-β1 did not derive from an increase in latent TGF-β1 . A commercial ELISA for TGF-β1 indicated that the TGF-β detected in the luciferase bioassay was TGF-β1 and not other TGF-β isoforms , and confirmed the induction of TGF-β1 in HCMV infected , raTGF-β1 infected cells ( Figure 3E ) . The quantity of active TGF-β1 protein measured in the ELISA was approximately 3 log ( 1000-fold ) lower than the quantity of TGF-β1 activity detected by luciferase assay . These studies were performed with equivalent volume samples obtained from the same experimental cultures , confirming that the picomolar measurements indeed differed between the two assays; however , significant induction of active TGF-β1 was measured in the HCMV infected , raTGF-β1 treated cells by both methods . These studies have not been performed in parallel by other investigators , and the differing results obtained from assays of the same samples suggests that the luciferase bioassay may detect the downstream function of a given quantity of protein more sensitively than the measurement of protein by antibody binding in the ELISA . Together , these results suggested that HCMV infection of HK-2 cells , without EMT , did not induce de novo production of active TGF-β1 , whereas the HCMV infected epithelial cells after EMT appeared to acquire the capacity to produce de novo active TGF-β1 , similar to the phenotype demonstrated by HCMV infected fibroblasts [25] . Uninfected HK-2 cells , before or after EMT , did not acquire the capacity to produce de novo active TGF-β1 . These studies were repeated using primary renal proximal tubular epithelial cells ( Figure 3C , D ) . These cells produced undetectable active TGF-β1 at baseline ( HCMV TR-/raTGF-β1- ) , with HCMV infection alone ( HCMV TR+/raTGF-β1- ) , and produced only a small amount of detectable active TGF-β1 after raTGF-β1 stimulation ( HCMV TR-/raTGF-β1+ ) , consistent with data derived from studies of HK-2 cells ( Figure 3A ) . Similar to the HK-2 cells , HCMV infected , raTGF-β1 stimulated primary renal epithelial cells ( HCMV TR+/raTGF-β1+ ) produced active TGF-β1 ( p<0 . 01 ) . The total TGF-β1 was similar in all treatment conditions , consistent with results from HK-2 cells ( Figure 3D ) . These results indicated that the induction of active TGF-β1 production observed in HCMV infected HK-2 cells after EMT was not unique to immortalized cells such as HK-2 but was likely a general phenotype observed for both immortalized and primary renal tubular epithelial cells after HCMV infection . Next , HK-2 cells were infected with HCMV at an MOI of 1 , stimulated with raTGF-β1 at 15 ng/ml , and incubated with a function blocking antibody against TGF-β1 in increasing concentrations from 0 . 3 µg/ml to 3 µg/ml ( Figure 3F ) . Increasing amounts of anti-TGF-β1 resulted in progressively decreasing quantities of active TGF-β1 produced . This effect was not due to the presence of anti-TGF-β1 in the luciferase assay because the blocking antibody was removed during the final washing , prior to incubation with fresh media for the luciferase assay . These results show that blockade of the stimulating dose of raTGF-β1 abrogated active TGF-β1 production . To determine whether the TGF-β1 blocking antibody might block effects of the TGF-β1 produced by HCMV infected cells , EMT was first induced in uninfected HK-2 cells using raTGF-β1 for 48 hours , after which cells were washed to remove raTGF-β1 , infected with HCMV at MOI of 1 , and finally incubated with either media alone or media containing the TGF-β1 blocking antibody at 3 µg/ml . After 24 hours , cells were lysed , RNA extracted and RT-PCR performed for mRNAs induced in the PCR array . Results from cells incubated with the blocking antibody are shown as percent reduction compared to cells incubated with media alone ( Figure S3 ) . These results show that cells , after undergoing EMT and subsequently infected with HCMV , demonstrate 50–95% reduction of mRNA transcripts of molecules associated with EMT in the presence of the TGF-β1 blocking antibody , compared to similar cells in the absence of the TGF-β1 blocking antibody . This result suggests that the TGF-β1 produced by HCMV infected cells after EMT may have activity upon the producing cells ( true autocrine activity ) , which is inhibited by the TGF-β1 blocking antibody . HCMV infected HK-2 cells were then stimulated with increasing amounts of raTGF-β1 at 0 , 1 , 3 , 5 , 10 , 15 , and 20 ng/ml , ( 0 . 04 , 0 . 12 , 0 . 2 , 0 . 4 , 0 . 6 , 0 . 8 nM ) and supernatants assayed for TGF-β1 activity using the luciferase bioassay . Active TGF-β1 increased with increasing stimulation dose ( Figure 4A ) , whereas total TGF-β1 production under all conditions remained similar ( dns ) . This result suggested that HCMV infected cells increased active TGF-β1 production in proportion to the quantity of TGF-β1 in the environment ( autocrine production or auto-induction ) [39] . Taken together , these results suggested that the quantity of TGF-β1 within an HCMV infected kidney might increase via autocrine production of additional TGF-β1 by HCMV infected epithelial cells undergoing EMT . To determine whether the number of HCMV infected cells could influence the amount of active TGF-β1 produced , HK-2 cells were infected with HCMV strain TR at increasing multiplicity of infection ( MOI ) ranging from 2 to 10 , stimulated with raTGF-β1 at 15 ng/ml ( 0 . 6 nM ) , and supernatants assayed for TGF-β1 production by luciferase bioassay . Increasing amounts of active TGF-β1 were produced by cells infected with HCMV at increasing MOI ( Figure 4B ) . These results suggested that the level of active TGF-β1 increased with multiplicity of HCMV infection . To determine whether TGF-β1 activation could be induced by HCMV strains other than TR , we utilized HCMV strain AD169 BADrUL131 ( kindly provided by T . Shenk , Princeton University , Princeton NJ ) , which contains a repaired mutation of the UL131 open reading frame ( ORF ) , permitting infection of epithelial cells [40] . Wild-type AD169 and other HCMV strains commonly utilized in vitro , such as Towne , contain mutations in ORF UL128-131 which prevent infection of epithelial cells [40] . BADrUL131 readily infected HK-2 cells as shown by immunofluorescent staining for IE1 antigen ( dns ) , and was used to infect HK-2 cells with or without raTGF-β1 stimulation , followed by luciferase bioassay for TGF-β1 in the supernatants ( Figure 4C ) . Similar to results for HCMV strain TR , BADrUL131 alone did not induce TGF-β1 production ( BADrUL131+/raTGF-β1- ) , whereas infected cells stimulated by raTGF-β1 ( BADrUL131 +/raTGF-β1+ ) did induce active TGF-β1 production ( p<0 . 01 ) . Thus , induction of active TGF-β1 production by HCMV infected renal epithelium after EMT did not appear to be HCMV strain-specific . To determine whether TGF-β1 production could be triggered by viral binding and entry into HK-2 cells without productive viral infection , HCMV strain TR was inactivated by ultraviolet irradiation ( HCMV TR , UV+ ) and used to infect HK-2 cells . HCMV uninfected cells ( HCMV TR - ) and non-irradiated HCMV infected cells ( HCMV TR + ) served as negative and positive controls , respectively . Under these conditions , irradiated HCMV did not induce active TGF-β1 production after EMT ( p>0 . 05 ) , and did not affect total TGF-β1 levels either before or after EMT ( Figure 4D ) . These results indicated that active TGF-β1 could not be induced solely by binding and entry of viral particles into cells undergoing EMT . To characterize the mechanism by which HCMV infection might induce active TGF-β1 production , we added known inhibitors of TGF-β1 activation - aprotinin ( serine protease inhibitor against plasmin ) , GM6001 ( matrix metalloprotease inhibitor ) , anti-thrombospondin-1 , anti-αvβ6 integrin - to HCMV infected HK-2 cells prior to raTGF-β1 stimulation and again after washing , and the TGF-β1 luciferase bioassay was performed ( Figure 5A ) . Active TGF-β1 production was reduced by approximately 60% ( p<0 . 01 ) in the presence of GM6001 . In the presence of aprotinin , active TGF-β1 production was decreased to a lesser but still statistically significant amount ( 19% , p<0 . 05 ) . Active TGF-β1 was not significantly decreased in the presence of the other inhibitors , the blocking antibodies against thrombospondin 1 and the αvβ6 integrin . This result suggested that MMPs and serine proteases might be involved in TGF-β1 activation by HCMV infected epithelial cells after EMT . Because TGF-β1 is activated by MMP-2 , we next analyzed MMP-2 production in our experimental conditions . Lysates from HK-2 cells with or without HCMV infection and/or raTGF-β1 stimulation were separated under non-reducing conditions by gelatin-containing SDS-PAGE and developed for presence of gelatinase activity . In Figure 5B , a representative zymogram showed bands at 72 and 62 kDa , consistent with pro- and active MMP-2 , only in lysates from HCMV infected , raTGF-β1 stimulated cells ( HCMV TR +/raTGF-β1 + ) . Identity of these bands as MMP-2 was confirmed by immunoblotting using anti-MMP-2 ( Figure 5B ) . This result indicated that only HCMV infected , raTGF-β1 stimulated cells had cell-associated pro-MMP-2 and active MMP-2 as detected in this assay of functional MMP-2 . Since pro-MMP-2 is known to undergo activation in complex with other MMPs on the cell surface , we next performed immunoprecipitation experiments using lysates of HK-2 cells with or without HCMV infection and/or raTGF-β1 stimulation . Lysates were subjected to western blotting for TIMP-2 , MT3-MMP , and MT1-MMP directly ( Figure 5C ) , or were incubated with mouse anti-MMP-2 , immune complexes collected with protein A-agarose , and Western blotting of immunoprecipitated proteins performed using a rabbit anti-MMP-2 antibody , or antibodies against TIMP-2 , MT3-MMP , and MT1-MMP ( Figure 5D ) . Immunoblotting for MMP-2 in the immunoprecipitates showed that MMP-2 was detectable in all samples after immunoprecipitation ( Figure 5D ) . Immunoblotting of immunoprecipitated proteins demonstrated the presence of TIMP-2 and MT3-MMP only in the HCMV infected , raTGF-β1 treated cells ( HCMV TR+/raTGF-β1+ ) ( Figure 5D ) . MT1-MMP did not immunoprecipitate with anti-MMP-2 in any samples , and was not detectable even with membrane overexposure ( Figure 5D ) . These results suggested that MMP-2 activation by HCMV infected , raTGF-β1 stimulated cells could occur via formation of a membrane-associated complex of MT3-MMP , TIMP-2 , and MMP-2 [41] . To determine whether reduction of MMP-2 production could inhibit TGF-β1 activation , we next transfected a GFP-expressing shRNA construct against MMP-2 mRNA into HK-2 cells prior to HCMV infection and raTGF-β1 stimulation . A control plasmid , consisting of scrambled RNA serving to control for off-target effects , was transfected in parallel cultures . GFP expression was confirmed by fluorescence microscopy daily during the assay and confirmed similar transfection efficiencies . Active TGF-β1 in supernatants was evaluated by luciferase assay , and cell lysates were divided into equal portions and either assessed by western blotting for MMP-2 , GFP , and actin , or RNA extracted for RT-PCR analysis of MMP-2 . Active TGF-β1 production was inhibited by the MMP-2 shRNA construct but not the control irrelevant shRNA ( Figure 5E , top panel ) . MMP-2 protein was not detectable in the samples transfected with the MMP-2 shRNA , but was detectable in the samples transfected with the control irrelevant shRNA treated with raTGF-β1 ( Figure 5E , middle panel ) . GFP and actin expression were similar in all samples ( Figure 5E , middle panel ) . RT-PCR analysis , depicted as a fold change between samples with and without raTGF-β1 exposure for either MMP-2 shRNA or the control shRNA , demonstrated detectable MMP-2 mRNA only in the control shRNA ( Figure 5E , bottom panel ) . Taken together , these results indicate that reduction of MMP-2 production results in inhibition of TGF-β1 activation in this experimental system . To characterize the viral gene product requirements for TGF-β1 activation by HCMV infected epithelial cells after EMT , viral polymerase inhibitors ganciclovir ( GCV ) and foscarnet ( PFA ) were added at a range of concentrations to inhibit viral replication , one hour after HCMV TR infection but before raTGF-β1 stimulation , and again after washing , and the luciferase bioassay for TGF-β1 performed ( Figure 6A , left panel ) . For each condition , DNA was extracted from cell lysates and quantitative DNA PCR for the HCMV UL55 ORF ( gB ) was performed ( Figure 6A , right panel ) . Production of active and total TGF-β1 was not affected by either of the viral inhibitors , suggesting that the viral effect resulting in TGF-β1 production preceded viral replication as represented by viral polymerase inhibition . Together with results from experiments using irradiated viruses , this result suggested that TGF-β1 production after EMT and HCMV infection might involve the function of immediate early or early gene products . To investigate the possible function of the major HCMV immediate early genes IE1 or IE2 in TGF-β1 activation , HK-2 cells were transiently transfected with expression plasmids containing the open reading frame for either IE1 or IE2 , stimulated with raTGF-β1 , and luciferase bioassay performed . A plasmid containing ORF UL55 , encoding gB , was also transfected separately into HK-2 cells as a representative HCMV late gene product . Expression of transfected constructs was confirmed by immunofluorescent analysis for HCMV IE1 , IE2 , or gB antigens by cells grown on coverslips and by western blotting for IE1 , IE2 , or gB proteins in transfected cell lysates ( dns ) . A lacZ-expressing plasmid was transfected separately or co-transfected with the other plasmids , and transfection efficiency was determined by quantitation of β-galactosidase activity in cell lysates . Results of the bioassay for active TGF-β1 were normalized to transfection efficiency . Plasmids containing either IE1 or IE2 did not induce TGF-β1 production before EMT , but did induce active TGF-β1 production after EMT , whereas control plasmids containing HCMV UL55 or lacZ did not induce TGF-β1 production either before or after EMT ( Figure 6B ) . This result suggested that IE1 and IE2 might have a common transactivating function permitting TGF-β1 production by HCMV infected HK-2 cells after EMT . Because MMP-2 was implicated in the TGF-β1 activating phenotype in HCMV infected cells ( Figure 5 ) , lysates from cells transfected with either IE1 , IE2 , or gB were divided equally and subjected to western blotting for MMP-2 or RT-PCR for MMP-2 mRNA . MMP-2 protein and mRNA were detectable in IE1 and IE2 transfections treated with raTGF-β1 , but not in gB tranfections ( Figure 6C , D ) . Immunoblotting for actin was performed as a loading control . These results are consistent with those from HCMV infected cells and suggest that the HCMV IE1/2 gene products may promote TGF-β1 activation via MMP-2 induction .
Epithelial-to-mesenchymal transition ( EMT ) is a well-characterized phenotypic change manifested by renal tubular epithelial cells after exposure to TGF-β1 and has been associated with various forms of renal fibrosis . TGF-β1 is present in renal allografts , including both human renal biopsies after transplantation , and in animal models of renal transplantation . TGF-β1 has been described to be more abundant in CMV infected renal allografts in both patient biopsies and animal transplantation models . However , the phenotype of CMV infected renal epithelial cells undergoing EMT has not previously been characterized . In this study we have shown that , in vitro , HCMV infected renal tubular epithelial cells can undergo EMT after exposure to TGF-β1 . After EMT , HCMV infected cells can induce extracellular activation of TGF-β1 via induction of MMP-2 . This autocrine production of TGF-β1 could contribute to the observation in human and animal renal transplantation associating HCMV infection with increased expression of intra-renal TGF-β1 . This autocrine process is thought to contribute to pathologic fibrosis in vivo [39] . Our in vitro findings are consistent with findings in vivo by Helantera et al . , showing increased TGF-β1 in HCMV infected human renal biopsies as well as increased urinary TGF-β1 from patients with HCMV infected allografts [15] , [42] . MMPs also participate in degradation of basement membrane , enhancement of cellular motility , activation of growth factors , and modulation of cell adhesion molecules , and have been implicated in various forms of renal fibrotic disease [43] , [44] . Elevation of urinary MMPs has been described during renal allograft rejection but has not been explored in the context of HCMV infected renal allografts [45] , [46] . Our finding that the complex of MMP-2 , MT3-MMP , and TIMP-2 may contribute to TGF-β1 activation in HCMV infected HK-2 cells demonstrates that this ternary complex , which has been well described in vitro , may serve as a mechanism for TGF-β1 activation by HCMV infected renal tubular epithelial cells [41] , [47] , [48] . The concept that other factors ( in this case , HCMV infection ) might amplify the fibrogenic phenotype of HK-2 cells undergoing TGF-β1 induced EMT has been validated by others , who have shown that epidermal growth factor enhances the migratory phenotype of HK-2 cells and synergistically increases MMP-9 production after TGF-β1 induced EMT [49] . Interestingly , the fibrogenic phenotype enhanced by HCMV occurs only after EMT , as HCMV infected HK-2 cells in epithelial form did not manifest significant induction of TGF-β1 . This suggests that HCMV infected renal epithelial cells , at baseline , do not induce fibrogenic renal changes , which would be consistent with the absence of primary renal pathology in asymptomatic humans infected with HCMV . CMV effects upon extracellular matrix and fibrosis have been previously characterized in the context of rat CMV infection of renal and cardiac allografts . In those models , rat CMV intensified production of fibrogenic molecules such as TGF-β1 , PDGF , and collagens in renal allografts , and was associated with transcriptional upregulation of numerous fibrogenic and angiogenic molecules in the cardiac allograft [7] , [17] , [50] . HCMV infection has also been shown to induce TGF-β1 activation by endothelial cells via an integrin-mediated mechanism , suggesting that placental infection by HCMV may alter extracellular matrix and permit HCMV translocation across the placenta , contributing to congenital infection of the fetus [31] . Our work , the first to investigate HCMV infection of renal tubular cells in vitro , supports the findings by others that HCMV may modify the extracellular matrix during inflammatory conditions such as solid organ transplantation or chorioamnionitis . We have also shown that , similar to fibroblasts and astrocytes , transient transfection of plasmids encoding the HCMV IE1 or IE2 gene products can induce TGF-β1 activation via MMP-2 by renal epithelial cells after EMT . The inability of viral DNA polymerase inhibitors , ganciclovir and foscarnet , to affect TGF-β1 activation after EMT in HCMV infected epithelial cells supports the role of gene products from IE1 and/or IE2 in this phenotype , and suggests potential limited utility of these antivirals in preventing HCMV associated fibrosis in the clinical setting . Interestingly , the promoters for IE1 , IE2 , and numerous MMPs all contain AP-1 binding sites , and in the case of MMPs , this transcription factor is thought to contribute to control of transcriptional upregulation [51] . Our finding that either the IE1 or IE2 gene products may be sufficient to induce TGF-β1 production suggests that HCMV reactivation within the transplanted kidney may be associated with the pathogenesis of HCMV associated renal allograft damage . These conditions would be present uniquely in renal allografts , with host T cell immunosuppression and local allograft inflammation permitting HCMV reactivation within the allograft , and presence of local TGF-β1 in the allograft inducing EMT [34] , [52] , [53] . These conditions would all exist simultaneously after renal transplantation , but not necessarily in other forms of fibrotic renal disease in which TGF-β1 may be present . In summary , this in vitro model shows that HCMV infected renal tubular epithelial cells undergo EMT after exposure to TGF-β1 . HCMV infection may amplify autocrine TGF-β1 production via an MMP cascade . The HCMV IE1/IE2 transcription factors are each capable of inducing active TGF-β1 production , suggesting that clinically utilized antivirals might not prevent this HCMV effect within the kidney . These in vitro studies provide a potential pathogenic mechanism for the observed association between HCMV infection , TGF-β1 production , and poor clinical allograft outcome in human renal transplant recipients .
The immortalized human renal proximal tubular epithelial cell line , HK-2 , was purchased from American Type Culture Collection ( Manassas , VA ) and maintained in keratinocyte serum-free media ( Invitrogen , Carlsbad CA ) . Primary human renal proximal tubular epithelial cells ( Lonza , Walkersville MD ) were maintained in renal epithelial growth media ( Lonza ) . Mink lung epithelial cells stably expressing the TGF-β response element of the plasminogen activator inhibitor-1 promoter fused to the firefly luciferase reporter gene ( gift of D . Rifkin , New York University , New York NY ) were maintained as described [54] . Primary human foreskin fibroblasts ( HFFs ) were maintained as described [55] . HCMV strain TR ( gift of J . Nelson , Oregon Health and Sciences University , Portland OR ) and BADrUL131 ( gift of T . Shenk , Princeton University , Princeton NJ ) were propagated in HFFs . Viruses were concentrated by centrifugation at 16 , 000×g for 2 hours at 4°C , resuspended in keratinocyte serum-free media , and frozen at −80°C until use . For ultraviolet inactivation , HCMV was exposed to ultraviolet radiation at 150 mJ in a cross-linking chamber ( Bio-Rad , Hercules CA ) [56] . Virus titers were determined using a standard assay for detection of early antigen fluorescent foci ( DEAFF ) in fibroblasts [57] , [58] . The following reagents were purchased from commercial vendors: recombinant human active TGF-β1 ( raTGF-β1 ) , Quantikine human TGF-β1 ELISA , TGF-β1 blocking antibody ( clone 9016 ) ( R&D Systems , Minneapolis MN ) ; luciferase assay reagent , β-galactosidase assay kit ( Promega Corp . , Madison WI ) ; RNeasy kit ( Qiagen , Valencia CA ) ; RT2 First Strand Kit , SuperArray Human Extracellular Matrix PCR Array , human MMP-2 shRNA kit ( SABiosciences , Frederick MD ) , Cells-to-CT kit ( Applied Biosystems , Foster City CA ) ; GM6001 , rabbit anti-MMP-2 antibody ( AB19167 ) ; rabbit anti- MT3-MMP antibody ( AB853 ) , mouse anti-αvβ6 blocking antibody ( MAB2077Z ) ( Millipore , Billerica MA ) ; pEF1/myc-his/lacZ plasmid , anti-GFP antibody , AlexaFluor conjugated phalloidin and secondary antibodies , and SuperScript III kit ( Invitrogen ) ; Cytogam ( CSL Behring , King of Prussia PA ) ; Nucleofector device and transfection kit V ( Amaxa , Gaithersburg MD ) . The following human primer/probe sets were purchased from ABI: fibronectin ( Hs . 01549976_m1 ) ; MMP-9 ( Hs . 00957562_m1 ) ; ADAMTS1 ( Hs . 00199608_m1 ) ; TGF-β1 ( Hs . 00932734_m1 ) ; collagen 5A1 ( Hs . 00609088_m1 ) ; MMP-2 ( Hs . 01548733_m1 ) ; thrombospondin-1 ( Hs . 00170236_m1 ) ; 18S RNA ( part #4333760-0904029 ) . Mouse anti-MMP-2 ( mab CA801 ) , mouse anti-TIMP-2 ( mab101 ) , and rabbit anti-MT1-MMP polyclonal antisera ( pab 198 ) were a kind gift from R . Fridman ( Wayne State University , Detroit MI ) . Ganciclovir and foscarnet were kindly provided by M . Prichard ( University of Alabama-Birmingham , Birmingham AL ) . Anti-thrombospondin 1 antibody ( mab133 ) , mouse monoclonal antibodies against HCMV IE1 , IE2 , and gB ( mab 63-27 , mab 2-9-5 , and mab 7-17 ) , and expression plasmids containing HCMV IE1 , IE2 , or UL55 open reading frames were provided by the authors ( J . M . -U . and W . B . ) . Primary HFFs , HK-2 cells , or primary renal tubular epithelial cells were grown in 6-well plates , infected with HCMV at an MOI of 1 for one hour , washed and incubated with fresh media , and cells and media harvested daily for 6 days and stored at −80°C . Cells were resuspended in 500 µl DPBS , sonicated briefly , serial dilutions made , and DEAFF assay performed . Serial dilutions of media were also analyzed by DEAFF assay . Cells were grown on coverslips , infected with HCMV TR and/or incubated with recombinant active TGF-β1 ( raTGF-β1 ) for 48 hours , fixed in 4% paraformaldehyde , and permeabilized using 0 . 1% Triton X-100 . Cells were incubated with primary antibodies overnight at 4°C , washed , incubated for 1 hour with isotype and species specific secondary antibodies labeled with either AlexaFluor 488 or AlexaFluor 594 followed by Topro3 nuclear stain and mounted using ProLong anti-fade reagent . Images were collected under similar exposure times and identical gain using confocal fluorescence microscopy ( Olympus Fluoview BX51 , Center Valley PA ) . HK-2 cells were untreated , or infected with HCMV TR and/or treated with raTGF-β1 at 15 ng/ml ( 0 . 6 nM ) . Cells were washed , pelleted by centrifugation , lysed , and equivalent protein lysates were boiled and subjected to 10% SDS-PAGE and immunoblotting with enhanced chemiluminescence as described [59] . Immunoblots were analyzed using Adobe Photoshop densitometry software ( Adobe Systems Inc , San Jose CA ) . For phospho-SMAD2 immunoblotting , cells were harvested at 1 hour after raTGF-β1 exposure , lysed and nuclear extracts prepared as described [60] . Nuclear proteins were separated by SDS-PAGE , transferred to nitrocellulose , and immunoblotting performed as above . HK-2 cells were grown in T75 flasks , either untreated or infected with HCMV TR and/or incubated with raTGF-β1 for 24 hours , lysed using TriPure reagent , and RNA extracted using the RNeasy kit . RNA was reverse transcribed to cDNA using the RT2 First Strand Kit , and cDNAs used according to the manufacturer's instructions for the SuperArray Human Extracellular Matrix PCR Array . Transcripts were normalized by comparison to GAPDH cDNA , and samples from experimental conditions were quantitated as fold-change compared to baseline production in uninfected , unstimulated HK-2 cells . For RT-PCR using commercial primer/probe sets ( ABI ) , HK-2 cells or primary renal tubular epithelial cells were prepared as described and RNA extracted using the RNeasy kit . RNA was reverse transcribed to cDNA using the SuperScript III Kit . Reactions were performed in triplicate in 20 µl volumes consisting of ABI 2x MasterMix ( 10 µl ) , ABI primer/probe mix ( 1 µl ) , template ( 1 µl ) , and water ( 8 µl ) and real-time PCR performed according to the manufacturer's instructions . Results were normalized to 18S RNA and depicted as fold change from baseline ( uninfected , unstimulated cells ) . Experiments were performed three separate times . HK-2 cells or primary renal tubular epithelial cells were seeded onto 24 well plates ( 2×104 cells/well ) or 6 well plates ( 1×105 cells/well ) , either untreated or infected with HCMV at an MOI of 1 ( strain TR , BADrUL131 , or UV inactivated TR ) and/or treated with raTGF-β1 for 48 hours to induce EMT . For the CMV MOI assay , HCMV strain TR was added at an MOI of 2 , 4 , 6 , 8 , or 10 to each well . Wells were then washed three times with media to remove all raTGF-β1 and re-incubated with fresh media not containing raTGF-β1 . Supernatants were collected at 24 hours post wash and stored at −80°C until assayed for TGF-β1 production . For the luciferase bioassay , supernatants were assayed for active and total TGF-β production using the mink lung epithelial cell reporter bioassay as described [54] . TGF-β concentrations ( ρmol/ml ) were calculated by comparison to a standard curve derived from known quantities of raTGF-β1 . The Quantikine human TGF-β1 ELISA and the Ebioscience ELISA were used to quantitate active and total TGF-β1 according to the manufacturer's instructions . Some studies were performed in parallel using samples from the same experiments for both luciferase assays and ELISA for direct comparison of both assays . In some experiments , HK-2 cells were untreated or infected with HCMV TR at MOI of 1 and/or stimulated with raTGF-β1 at 15 ng/ml ( 0 . 6 nM ) . The TGF-β1 blocking antibody was added at 0 . 3 , 1 , 2 , and 3 µg/ml prior to addition of raTGF-β1 but was not re-added after washing and addition of fresh media for luciferase assay . Next , HK-2 cells were stimulated to undergo EMT by exposure to raTGF-β1 for 48 hours . Cells were washed three times with media to remove exogenous raTGF-β1 , then were infected with HCMV at MOI of 1 . Cells were either incubated with media alone , or with media containing the TGF-β1 blocking antibody at 3 µg/ml for 24 hours . Cells were washed , lysed and total RNA extracted using the RNeasy kit . Reverse transcription assays were performed as described , and cDNA used as template for real-time PCR assays as described . Results from samples incubated with the TGF-β1 blocking antibody were compared to those from samples without the blocking antibody ( baseline ) , and differences in mRNA expression depicted as percent reduction from baseline . HK-2 cells were untreated or infected with HCMV TR at MOI of 1 and/or stimulated with raTGF-β1 at 15 ng/ml ( 0 . 6 nM ) . TGF-β1 inhibitors were added to wells one hour prior to raTGF-β1 stimulation at the following concentrations: GM6001 at 0 . 5 nM; aprotinin at 200 µg/ml; anti-thrombospondin-1 at 25 ng/ml; anti-αvβ6 at 10 µg/ml [61] , [62] , [63] , [64] . After 48 hours , cells were washed , inhibitors re-added , supernatants harvested at 24 hours post-wash , and luciferase bioassay performed . HK-2 cells were untreated or infected with HCMV TR at MOI of 1 and/or treated with raTGF-β1 at 15 ng/ml ( 0 . 6 nM ) . Cells were washed , pelleted by centrifugation , lysed , and equivalent protein lysates were subjected to gelatin zymography as described [65] . Gels were stained with Coomassie Blue and photographed using the Quantity One gel imaging system ( Bio-Rad ) . HK-2 cells were untreated , or infected with HCMV TR at MOI of 1 and/or stimulated with raTGF-β1 as described . Cells were washed , pelleted by centrifugation , lysed in cold RIPA buffer with protease inhibitors , incubated at 4°C overnight with anti-MMP-2 , followed by incubation with protein A-agarose , washed and resuspended in RIPA buffer , boiled , and separated by 8% SDS-PAGE under reducing conditions . An aliquot of the original cellular lysate was saved prior to immunoprecipitation . Western blotting was performed for the original lysate and the immunoprecipitated material as described using antibodies against TIMP-2 , MT1-MMP , and MT3-MMP . HK-2 cells were transfected with a commercial MMP-2 shRNA construct or a control scrambled construct provided by the manufacturer ( SABiosciences ) , using the Amaxa Nucleofector kit V and Nucleofector program T-020 according to the manufacturer's instructions . Tranfected cells were each divided into two separate wells and observed daily by fluorescence microscopy for GFP expression . Wells were infected with HCMV strain TR at MOI of 1 , and one of each pair were stimulated with raTGF-β1 , washed , and supernatants and cell pellets harvested . Supernatants were assayed for TGF-β1 production by luciferase assay . Cell pellets were divided into equal portions , half of which was subjected to western blotting for MMP-2 , GFP , and actin . The other half underwent total RNA extraction using the Cells to CT kit , and RT-PCR was performed using primer/probe sets for MMP-2 and 18S RNA according to the manufacturer's instructions ( ABI ) . Results were normalized to 18S mRNA expression and depicted as fold change between samples treated with and without raTGF-β1 for each shRNA transfection condition . HCMV TR infected HK-2 cells on coverslips were incubated with ganciclovir at 50 , 100 , 200 , and 300 µM , foscarnet at 112 . 5 , 325 , 500 , and 1000 µM , stimulated with raTGF-β1 at 15 ng/ml ( 0 . 6 nM ) , washed and re-incubated with inhibitors , supernatants harvested 24 hours after washing and luciferase bioassay performed [66] , [67] , [68] . Cells were harvested separately , DNA extraction performed using the Qiagen DNA Blood kit , and quantitative DNA PCR performed using primers and probe for a highly conserved sequence within HCMV gB and compared to a standard curve generated by real-time PCR amplification of known copy numbers ( 101–108 copies ) of a plasmid containing the gB DNA sequence recognized by the gB primer/probe pair [69] . Results were depicted as DNA copies based upon comparison with the plasmid standard curve . In 6-well plates containing coverslips , 1×106 HK-2 cells were transfected with the expression plasmid containing lacZ either alone or with expression plasmids containing either IE1 , IE2 , or UL55 , using the Nucleofector kit V and Nucleofector program T-020 . At 24 hours post transfection , some wells were stimulated with raTGF-β1 at 15 ng/ml ( 0 . 6 nM ) for 48 hours , washed , and all supernatants assayed for TGF-β1 by luciferase bioassay . Coverslips were fixed and stained as described for IE1 , IE2 , or gB protein expression . Cells were lysed and assayed for β-galactosidase activity according to the manufacturer's instructions using an ELISA microplate reader ( ELx808 , Bio-Tek Instruments , Inc . , Winooski , VT ) . Results from the luciferase bioassay were normalized to relative transfection efficiency as determined by β-galactosidase expression . In a separate experiment , transfections and raTGF-β1 stimulation were performed as described , and cell lysates were separated into two aliquots and subjected either to western blotting for MMP-2 protein or RT-PCR for MMP-2 mRNA expression as described . Results were depicted as a fold change for each transfection condition between unstimulated and raTGF-β1 stimulated cells . All assays were performed with triplicate samples and results expressed as mean ± SEM . The Student T test and one-way analysis of variance ( ANOVA ) were used to compare groups using Prism 3 . 0 software , accepting statistically significant differences at a p value of<0 . 05 ( GraphPad , San Diego CA ) . All experiments , except the PCR array , were performed at least three times independently to confirm the reproducibility of each result . | Human cytomegalovirus ( HCMV ) is a common virus that establishes lifelong persistence in the host . Although asymptomatic in healthy people , HCMV can reactivate and cause disease in immunosuppressed patients , such as those undergoing kidney transplantation . HCMV infection is associated with inferior renal allograft survival compared to transplants without HCMV infection . HCMV infected allografts also contain higher levels of the fibrogenic cytokine , transforming growth factor-β1 ( TGF-β1 ) , compared to uninfected allografts . TGF-β1 is a potent inducer of renal fibrosis and causes epithelial-to-mesenchymal transition ( EMT ) , whereby epithelial cells acquire characteristics of cells of mesenchymal origin and express molecules associated with fibrosis . Our work shows that renal epithelial cells infected in vitro with HCMV can undergo EMT , but that HCMV infected cells produce greater amounts of the fibrogenic molecule TGF-β1 , compared to uninfected cells after EMT . We have shown that this effect is likely due to specific HCMV genes ( IE1 , IE2 ) , and cannot be prevented by administration of antivirals such as ganciclovir or foscarnet . These data suggest that HCMV may contribute to adverse renal allograft outcome by exacerbating TGF-β1 induced renal fibrosis . Understanding such mechanisms will permit the development of treatments that could improve long-term renal allograft survival in HCMV infected patients . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"virology",
"infectious",
"diseases/viral",
"infections",
"nephrology/dialysis",
"and",
"renal",
"transplantation"
] | 2010 | Human Cytomegalovirus Induces TGF-β1 Activation in Renal Tubular Epithelial Cells after Epithelial-to-Mesenchymal Transition |
Interferon stimulated genes ( ISGs ) target viruses at various stages of their infectious life cycles , including at the earliest stage of viral entry . Here we identify ArfGAP with dual pleckstrin homology ( PH ) domains 2 ( ADAP2 ) as a gene upregulated by type I IFN treatment in a STAT1-dependent manner . ADAP2 functions as a GTPase-activating protein ( GAP ) for Arf6 and binds to phosphatidylinositol 3 , 4 , 5-trisphosphate ( PI ( 3 , 4 , 5 ) P3 ) and PI ( 3 , 4 ) P2 . We show that overexpression of ADAP2 suppresses dengue virus ( DENV ) and vesicular stomatitis virus ( VSV ) infection in an Arf6 GAP activity-dependent manner , while exerting no effect on coxsackievirus B ( CVB ) or Sendai virus ( SeV ) replication . We further show that ADAP2 expression induces macropinocytosis and that ADAP2 strongly associates with actin-enriched membrane ruffles and with Rab8a- and LAMP1- , but not EEA1- or Rab7- , positive vesicles . Utilizing two techniques—light-sensitive neutral red ( NR ) -containing DENV and fluorescence assays for virus internalization—we show that ADAP2 primarily restricts DENV infection at the stage of virion entry and/or intracellular trafficking and that incoming DENV and VSV particles associate with ADAP2 during their entry . Taken together , this study identifies ADAP2 as an ISG that exerts antiviral effects against RNA viruses by altering Arf6-mediated trafficking to disrupt viral entry .
The induction of innate immune signaling is critical for host defense against viral infections , and is most commonly initiated by the detection of foreign nucleic acids as non-self . Once this system is activated , host cells orchestrate an array of signaling pathways that culminate in the induction of type I interferons ( IFNs ) , which include IFNα and IFNβ . Type I IFNs themselves possess no antiviral activity and instead exert their potent antiviral effects by the induction of hundreds of interferon-stimulated genes ( ISGs ) that can be induced by secreted IFNs in either an autocrine or paracrine manner . ISGs can function in a pan-viral manner or can target specific virus species and/or families [1] . Although the mechanisms by which several ISGs function to suppress viral infections have been well documented , the functions of many ISGs remain largely undefined . ISGs function to restrict viral replication at various stages of the viral life cycle , with some ISGs targeting the earliest event associated with infection—viral entry into the host cell . Some of the best-characterized ISGs that target viral entry belong to the interferon-inducible transmembrane protein ( IFITM ) family , which includes IFITM1 , IFITM2 , IFITM3 , and IFITM5 ( reviewed in [2 , 3] ) . IFITM family members exert broad antiviral effects against a diverse range of viruses including influenza A virus [4] , HIV [5] , dengue virus ( DENV ) [4] , and Ebola virus [6] . Although expressed at basal levels in many cell types , IFITM family members are induced by type I IFNs and may inhibit viral entry by direct alterations of cellular cholesterol homeostasis [7] , by alterations in the fusion between vesicular compartments that may favor a non-infectious entry pathway [8] , and/or by preventing fusion between viral and host-derived membranes while still permitting hemifusion [8 , 9] . The targeting of viruses at the earliest stages of their infectious life cycles serves as a potent step at which ISGs can antagonize viral infections . Viruses enter host cells through a variety of mechanisms . In some cases , viruses enter cells through similar means across multiple cell types , whereas in others , they utilize cell type-specific mechanisms for their entry . For example , vesicular stomatitis virus ( VSV ) exclusively utilizes a clathrin-mediated pathway for its entry [10 , 11] , whereas the enterovirus coxsackievirus B ( CVB ) enters via a dynamin II GTPase-independent pathway in polarized intestinal cells [12] , but enters nonpolarized epithelial cells and polarized epithelial cells through a dynamin II-mediated pathway [13 , 14] . Some viruses , such as Sendai virus ( SeV ) fuse directly at the host cell plasma membrane [15] . In the case of viruses that infect across diverse species , such as DENV that gains entry into both mosquito and human cells during its life cycle , viruses also exhibit species-specific entry mechanisms . In mosquito cells , DENV has been proposed to both enter via a clathrin-mediated pathway that requires delivery of incoming virions to a low pH endolysosomal compartment for fusion [16] , and possibly also by direct fusion at the plasma membrane [17] . In human cells , data support an uptake pathway for DENV that requires both clathrin coated pits and delivery of incoming virions to late endosomes [16 , 18] . Cellular factors that restrict virus entry may thus exhibit virus and cell-type specificity depending on the mechanism by which the virus gains entry into the host cell and/or mediates its uncoating and/or fusion . The process of endocytosis is tightly regulated by diverse cellular factors that specifically control the complex stages of the endocytic process . Many of these factors are GTP-binding proteins whose activity is regulated by both guanine nucleotide exchange factors ( GEFs ) and GTPase-activating proteins ( GAPs ) . Members of the ADP-ribosylation factor ( Arf ) family of Ras-related proteins are expressed in all eukaryotic cells and regulate both secretory pathway trafficking ( as is the case for Arf1 ) as well as cortical actin rearrangements and endocytosis ( as is the case for Arf6 ) . Arf6 influences cell migration , polarization , endocytosis , and endosomal trafficking via its direct impact on both regulators of the actin cytoskeleton [19 , 20] and the lipid membrane [21 , 22] . Thus , the expression of regulators of Arf6 activity , such as Arf6 GEFs and GAPs , can induce pronounced effects on a multitude of cellular events converging on uptake from the plasma membrane and/or endosomal trafficking . To identify additional ISGs that might directly impact virus entry and/or intracellular trafficking , we performed microarray analyses in control and signal transducer and activator of transcription ( STAT ) -1 signaling deficient cells exposed to purified IFNβ . Using this approach , we identified ArfGAP with dual pleckstrin homology ( PH ) domains 2 ( ADAP2 ) as a gene upregulated by IFNβ exposure in a STAT1-dependent manner . ADAP2 ( also known as centaurin-α2 , CENTA2 ) is a phosphatidylinositol 3 , 4 , 5-trisphosphate ( PI ( 3 , 4 , 5 ) P3 ) and PI ( 3 , 4 ) P2 binding protein whose expression alters Arf6 membrane localization [23–25] . However , ADAP2 has not been fully characterized and relatively little is known regarding its role in endocytosis or endosomal trafficking . Here we show that expression of ADAP2 suppresses DENV and VSV infection in an Arf6 GAP activity-dependent manner , while exerting no effect on CVB or SeV replication . We further show that expression of ADAP2 induces pronounced effects on the actin cytoskeleton and that it directly associates with actin-enriched membrane ruffles , macropinosomes , and lysosomes . Utilizing two techniques—a light-sensitive neutral red ( NR ) -containing DENV and fluorescence assays for virus internalization-—we show that ADAP2 primarily restricts DENV and VSV infection at the stage of virion entry or trafficking . Taken together , this study identifies an ISG that exerts its effects on DENV replication by altering Arf6-mediated trafficking to disrupt viral entry/trafficking .
To identify ISGs that might impact events associated with virus entry , we performed microarray analysis in control ( 2fTGH ) human fibrosarcoma HT1080 cells or cells lacking functional STAT1 ( U3A ) treated with purified IFNβ . As expected , we found that many known ISGs , such as interferon-induced protein 44-like ( IFI44L ) , members of the interferon-induced protein with tetratricopeptide repeats ( IFIT ) family , radical S-adenosyl methionine domain containing 2 ( RSAD2 ) , and members of the 2' , 5'-Oligoadenylate synthetase ( OAS ) family were upregulated by IFNβ treatment in a STAT1-mediated manner ( Fig 1A and S1 Table ) . In addition , we noted that the expression of ADAP2 , which contains Arf6 GAP activity and associates with actin cytoskeletal rearrangements , was also enhanced by IFNβ treatment in a STAT1-depdendent manner both by microarray ( Fig 1A ) and in follow up studies by RT-qPCR ( Fig 1B ) . Given the association of Arf6 with events that might impact viral entry , we chose to characterize both the antiviral and cell biological properties of ADAP2 . To determine whether ADAP2 exerts antiviral activity , we assessed the effects of its ectopic expression on the replication of DENV , VSV , CVB and SeV . We found that ADAP2 expression partially restricted DENV and VSV infection , but had no consistent effect on CVB or SeV infection ( Fig 1C ) . Expression of ADAP2 was confirmed by RT-qPCR ( S1A Fig ) . We also found that ADAP2 expression restricted DENV infection across a variety of multiplicity of infections ( MOIs ) ( S1B Fig ) . In addition , we found similar antiviral effects when EGFP-fused ADAP2 was ectopically expressed ( S1C and S1D Fig ) . Although overexpression of ADAP2 restricted DENV and VSV infection , we found that silencing of ADAP2 led to a mild increase in DENV infection ( S1H Fig ) . We attribute this result to the hundreds of other ISGs that are induced in response to viral infection and work in parallel to suppress viral infections . To exclude any impact of antiviral signaling on the restriction of DENV replication by ADAP2 , we next assessed the impact of ADAP2 overexpression on DENV and VSV replication in 293T cells depleted of the RIG-I-like receptor ( RLR ) adaptor mitochondrial antiviral signaling ( MAVS ) by CRISPR Cas9 genome editing . We found that expression of ADAP2 restricted DENV and VSV replication in both control cells and in cells lacking the expression of MAVS ( Fig 1D ) , supporting a RIG-I-like receptor ( RLR ) -independent pathway in the antiviral effects of ADAP2 . Expression of ADAP2 was confirmed by RT-qPCR ( S1E Fig ) . In addition , expression of ADAP2 has no impact on the induction of ISGs during DENV infection and instead reduced the induction given the inhibition of viral replication ( S1F and S1G Fig ) . Taken together , these data implicate a non-IFN-based mechanism of antiviral activity of ADAP2 . Relatively little is known regarding the localization and trafficking of ADAP2 . Therefore , we defined the localization of ADAP2 using real-time fluorescence microscopy in cells transfected with GFP-fused ADAP2 over the course of 48hrs post-transfection . Utilizing this approach , we found that expression of ADAP2 induced pronounced membrane ruffling and that it associated with both membrane ruffles and the vesicles that internalized from these ruffles ( Fig 2A , S1 Movie ) . Unfortunately , immunostaining with all commercially available ADAP2 antibodies was unsuccessful , thus we proceeded to characterize the cell biological localization of ADAP2 with ectopically expressed fusion proteins . To determine whether the ruffles induced by ADAP2 expression were actin-enriched , we also performed extended real-time fluorescence microscopy in cells transfected with YFP-fused actin and DsRed-fused ADAP2 . We found that DsRed-ADAP2 was associated with YFP-actin both at membrane ruffles and with vesicles internalizing from these ruffles ( Fig 2B , S2 Movie ) . In addition , we noted that in approximately 5–10% cells , ADAP2 overexpression induced the formation of very large actin-coated vacuoles that accumulated in large clusters ( Fig 2C ) . We next performed single vesicle-tracking analyses in cells expressing ADAP2 over an extended period ( ~4hrs ) to determine the pattern of movement exhibited by ADAP2-containing vesicles ( S3 Movie ) . We found that ADAP2-containing vesicles exhibited three distinct patterns of movement— ( 1 ) vesicles originating at the cell periphery/membrane ruffles that quickly internalized to the perinuclear region ( Fig 2D , pink tracings ) , ( 2 ) preformed vesicles that exhibited free movement within the cell ( Fig 2D , dark blue tracings ) , and ( 3 ) vesicles that were largely stationary and exhibited restricted movements ( Fig 2D , light blue tracings ) . Taken together , these data implicate ADAP2 in the enhancement of actin-enriched membrane ruffling . A hallmark of internalized macropinosomes is their association with actin . Given that we found that ADAP2 associated with actin-enriched membrane ruffles and internalized vesicles , we next assessed whether these vesicles were macropinosomes . To do this , we analyzed the localization of ADAP2 with fluorescently-labeled dextran , which serves as a marker for fluid-phase uptake . We found that expression of ADAP2 not only enhanced the extent of dextran uptake , but that ADAP2-containing cytoplasmic vesicles were enriched in dextran ( Fig 3A ) . Next , we determined the impact of expression of wild-type or constitutively-active ( Q111L ) and dominant-negative ( T66N ) mutants of Rab34 , a GTPase that has been specifically associated with macropinocytosis [26] , on the ability of ADAP2 to induce vesicles . We found that ADAP2 associated with Rab34-containing vesicles when both wild-type ( Fig 3B and 3C , S4 Movie ) and Q111L mutant ( Fig 3B and 3C ) were expressed . In addition , we found that expression of Rab34 Q111L enhanced the numbers and sizes of ADAP2-containing vesicles whereas expression of T66N Rab34 significantly reduced these vesicles and induced the accumulation of ADAP2 at cell-cell contacts and membrane ruffles ( Fig 3B–3F ) . Further supporting a role for macropinocytosis in the formation of ADAP2-enriched vesicles , we found that treatment of cells with 5- ( N-ethyl-N-isopropyl ) -Amiloride ( EIPA ) , an inhibitor of macropinoctyosis [27] , led to the accumulation of ADAP2 at the cell periphery and inhibited the formation of ADAP2-enriched vesicles ( Fig 3G ) . Taken together , these data support a role for ADAP2 in the induction of macropinocytosis , or a macropinocytosis-like process , in a Rab34-mediated manner . Rab8a GTPase localizes to membrane ruffles and macropinosomes as well as to tubular recycling endosomes that lead to membrane delivery back to the cell membrane [28 , 29] . Given that Rab8a localizes to macropinosomes , and that Arf6 functions upstream of Rab8a in this pathway [28] , we next determined whether ADAP2 associates with Rab8a . We found that Rab8 and ADAP2 exhibited significant colocalization in intracellular vesicles both in fixed ( Fig 4A ) and living ( Fig 4B and 4E , S5 Movie ) cells . In addition , we found that expression of a constitutively active mutant of Rab8a ( Q67L ) also highly associated with ADAP2-containing vesicles ( Fig 4C , S6 Movie ) and enhanced the numbers of total ADAP2 vesicles ( Fig 4D ) . As we detected ADAP2 at membrane ruffles and within intracellular cytoplasmic vesicles , these data suggest that ADAP2 localizes to Rab8a-positive recycling endosomes in addition to macropinosomes . Once internalized , conventional macropinosomes often fuse with one another , but generally exhibit little fusion with endosomes or lysosomes [30] . However , in some circumstances , internalized macropinosomes fuse directly with tubular lysosomes following their maturation [31] . To define the maturation process of ADAP2-containing macropinosomes , we assessed the localization of GFP-ADAP2 with markers of early endosomes ( early endosome antigen-1 ( EEA1 ) ) , late endosomes ( Rab7 ) , and lysosomes ( LAMP1 ) . We found that whereas ADAP2-containing vesicles were excluded from EEA1- and Rab7-containing vesicles , they exhibited a strong association with LAMP1-containing vesicles in both fixed ( Fig 5A–5D ) and living ( Fig 5E–5G , S7–S9 Movies ) cells . These data indicate that upon their internalization , ADAP2-containing vesicles strongly associate with LAMP1-positive lysosomes . ADAP2 contains an Arf6 GAP domain and two PH domains . We next determined whether the Arf6 GAP activity of ADAP2 was required for its induction of macropinocytosis . To do this , we constructed ADAP2 mutants in which the entire Arf6 GAP domain was deleted ( ΔARF GAP ) or the Arf6 GAP activity was abrogated by mutagenesis ( R53Q , a mutant described previously [23] ) ( Schematic , Fig 6A ) . Removal of the entire Arf6 GAP domain led to the appearance of largely immobile intracellular vesicles as analyzed by real-time fluorescence microscopy ( Fig 6B , S10 and S11 Movies ) and inhibited the ADAP2-mediated enhancement of macropinocytosis ( Fig 6C ) . In addition , an Arf6 GAP activity mutant of ADAP2 ( R53Q ) did not induce cytoplasmic vesicles and exhibited a strong association with the cell periphery ( Fig 6B , S12 Movie ) and did not induce macropinocytosis ( Fig 6C ) . Because we found that the Arf6 GAP activity of ADAP2 was required for its induction of macropinocytosis , we next assessed whether expression of Arf6 would impact ADAP2-mediated vesicle induction and whether the Arf6 GAP activity of ADAP2 was required for its vesicle induction . We found that expression of wild-type ADAP2 with wild-type Arf6 enhanced the formation of ADAP2-containing vesicles , although these vesicles were devoid of Arf6 ( S2A and S2B Fig ) . As the Arf6 GAP activity of ADAP2 was required for its induction of macropinocytosis , we next assessed whether this activity was also required for its restriction of DENV and VSV replication . We found that whereas expression of wild-type V5- or GFP-fused ADAP2 restricted DENV replication , expression of either the R53Q or ΔARF GAP mutants of ADAP2 had no significant effect on viral replication ( Fig 6D ) . Similarly , we found that expression of R53Q-ADAP2 had no effect on VSV infection ( Fig 6E ) . In both cases , transfection efficiency was verified by RT-qPCR ( S3A–S3C Fig ) . Taken together , these data implicate ADAP2 in the induction of marcopinocytosis and restriction of DENV and VSV replication via its Arf6 GAP activity . DENV gains entry into mammalian cells via a clathrin-mediated pathway that delivers incoming viral particles to late endosomes [16 , 18] . Likewise , VSV also enters cells via a clathrin-mediated pathway [10] and undergoes uncoating in early endosomes [32] . We found that expression of ADAP2 induced macropinocytosis through its Arf6 GAP activity , which was also required to restrict DENV and VSV replication . Given that DENV enters cells through a clathrin pathway , we next determined whether ADAP2 expression would alter clathrin-mediated endocytosis and/or DENV and VSV entry . We found that expression of wild-type ADAP2 , but not R53Q ADAP2 , altered the internalization of transferrin and prevented its perinuclear accumulation ( S4A and S4B Fig ) , suggesting that it alters cargo internalizing via the clathrin pathway . To determine if ADAP2 restricts DENV infection at the stage of viral entry , we first generated neutral red ( NR ) containing DENV particles ( DENV-NR ) . NR is an RNA-binding dye that has been used extensively in the field of picornavirus entry to identify inhibitors of virus entry [33–37] , but has not been used previously in the field of flavivirus entry . When viruses are propagated in the presence of NR , the dye associates with vRNA and renders the resulting NR-containing viral particles sensitive to light . Upon virus entry and RNA release , the NR dye diffuses away from the vRNA and replication continues in a light-insensitive manner . Thus , this method is a useful tool to establish the kinetics of viral entry and/or identify agents that inhibit and/or alter the release of vRNA . By propagating DENV in the presence of NR , we successfully generated light sensitive viral particles as confirmed by a greater than two-log drop in titer when fluorescent focus forming assays were performed under illuminated conditions ( Fig 7A ) . To determine whether ADAP2 restricts DENV entry , we performed a modified neutral-red infectious center ( NRIC ) assay [33] using DENV-NR particles ( a schematic of this assay is shown in Fig 7B ) . If ADAP2 restricts DENV entry , we would expect that exposure of ADAP2-expressing cells to light early in infection ( 2hrs p . i . ) , would elicit a greater inhibition of DENV infection than when cells are infected under non-illuminated conditions . Using this assay , we found that expression of ADAP2 inhibited DENV-NR replication under both non-illuminated and illuminated ( at 2hrs p . i . ) conditions ( Fig 7C ) . However , importantly , we found that the restriction of DENV-NR replication by ADAP2 was significantly enhanced under the illuminated ( at 2hrs p . i . ) condition ( Fig 7C ) , supporting a role for ADAP2 in the restriction of DENV entry . Next , we assessed whether incoming DENV and VSV associated with ADAP2 during their entry , as would be expected if it directly impacted their entry/trafficking . Similar to a previous study [18] , we found that DENV associated with early endosomes ( as assessed by its association with early endosome antigen-1 ( EEA1 ) ) early in its entry ( Fig 7D and 7G ) . However , we found that when ADAP2 was expressed , it exhibited a strong association with DENV particles early in its entry ( <30min p . i . ) ( Fig 7E and 7G ) . Similar to our findings with DENV , we found that ADAP2 also associated with VSV during its entry ( Fig 7F and 7G ) . As we have shown that ADAP2 does not associate with early endosomes , these DENV and VSV vesicles likely represent macropinosomes and/or lysosomes and suggest that ADAP2 alters the endocytic uptake of viral particles .
Inhibition of viral entry by ISGs is an effective strategy utilized by host cells to inhibit viral infection at the earliest stages of the viral life cycle . Here we show that ADAP2 expression is induced by type I IFNs in a STAT1-dependent manner and restricts DENV and VSV replication using an Arf6 GAP-mediated pathway , likely via the induction of macropinocytosis . Thus , our results point to a previously uncharacterized ISG that restricts DENV and VSV entry to limit viral replication . ADAP2 is ubiquitously expressed , with the highest levels of expression in the fat , heart , and skeletal muscle [38 , 39] . Unlike ADAP2 , the expression of the related ADAP1 ( also known as centaurin-α1 ) is largely restricted to the brain [25 , 40 , 41] . Although ADAP2 and ADAP1 share ~60% sequence identity and contain N-terminal ARF GAP domains and two PH domains , ADAP1 contains a nuclear localization signal and localizes primarily to the cytosol or nucleus whereas ADAP2 localizes predominantly to membrane ruffles . Sequence analysis of the promoter sequences of ADAP2 and ADAP1 revealed the presence of a conserved IFN-stimulated response element ( ISRE ) in the promoter of ADAP2 , but not in the promoter of ADAP1 ( S5 Fig ) . The induction of genes in response to type I IFNs is controlled by the presence of a conserved promoter sequence ( GAAA ( N ) GAAA , where N is any nucleotide ) termed the ISRE . Indeed , ADAP2 has been shown to be upregulated by IFN treatment in various human cells , such as primary hepatocytes [42] and PBMCs isolated from multiple sclerosis patients undergoing IFN treatments [43] . Taken together , our findings thus suggest that ADAP2 is specifically upregulated in response to type I IFN signaling , likely due to the presence of an ISRE . Our data suggest that ADAP2 inhibits DENV and VSV replication at the stage of viral entry . DENV enters mammalian cells by a clathrin-mediated endocytic pathway that delivers incoming viral particles to Rab7 positive late endosomes [16 , 18 , 44] . Although a small portion ( <20% ) of viral particles undergo membrane fusion in Rab5 early endosomes or Rab5/Rab7 intermediate endosomes , the majority ( >80% ) of membrane fusion occurs in Rab7 late endosomal compartments [35] . Thus , the delivery of incoming DENV particles to a late endosomal compartment is required for efficient membrane fusion and perturbation of this pathway could dramatically impact subsequent viral replication . Similarly , VSV enters cells via a clathrin-dependent pathway and requires Rab5-mediated delivery to an endosomal compartment ( with a pH of 6 . 2 ) to trigger G-protein-mediated fusion [11 , 45 , 46] , which is required to deliver particles into the host cell cytoplasm [32] . We show that expression of ADAP2 robustly induces macropinocytosis and that ADAP2-positive vesicles associate with Rab8a and the lysosomal marker LAMP1 , but not Rab7 or Rab5 . Our results therefore suggest that the induction of ADAP2 would serve to enhance macropinocytosis in cells exposed to type I IFN and that this would alter the trafficking of incoming DENV and VSV particles to preclude their delivery to Rab7 late endosomes . Moreover , the association of ADAP2-positive vesicles with LAMP1-positive compartments would serve to deliver any cargo , including viruses , contained within these vesicles to the degradative environment of lysosomes . Similar to our findings with DENV , we found that expression of ADAP2 partially restricted VSV infection . In contrast , ADAP2 expression had no effect on CVB or SeV replication . Unlike both DENV and VSV , CVB enters nonpolarized cells via a clathrin-independent pathway and does not require specific delivery to endosomal compartments for uncoating [14] . In addition , SeV fusion occurs at the host cell surface and does not rely on specific delivery to endosomal compartments for its entry . Thus , the differential effects of ADAP2 on viruses that require delivery to an endosomal compartment ( DENV and VSV ) and those that do not ( CVB , SeV ) support a role for ADAP2 in the mislocalizing of incoming virions to non-endosomal compartments , thus preventing their uncoating/fusion . This is supported by our findings that ADAP2 expression also led to the mislocalization of transferrin , which is specifically internalized by a clathrin-mediated pathway and delivered to the endosomal network . Consistent with this , we also found that expression of wild-type or Q67L Rab8a and wild-type or Q111L Rab34 , all of which induce macropinocytosis , restrict DENV infection ( S6 Fig ) . We found that both the anti-DENV and VSV and macropinocytosis-inducing effects of ADAP2 required its Arf6 GAP activity . The cycling of Arf6 between GTP and GDP bound states is a primary determinant for its impact on actin cytoskeletal dynamics and endosomal recycling . For example , expression of a constitutively active Arf6 mutant ( Q67L ) induces the accumulation of clathrin cargo in intracellular endosomal compartments , likely due to alterations in endosomal fusion [47 , 48] . In contrast , expression of a dominant inactive mutant ( T157N ) has no effect on intracellular accumulation of cargo , but instead exhibited overall lower levels of signal , presumably due to increases in endosomal recycling [49] . These data suggest that in cells expressing ADAP2 , Arf6 is maintained in a GDP-bound state , leading to both increases in macropinocytosis and endosomal recycling . Collectively , the promotion of these pathways would induce dramatic alterations in the uptake and intracellular trafficking of viral particles . Consistent with this , we found that expression of wild-type Arf6 had no effect on DENV infection whereas expression of Q67L Arf6 , which dramatically enhances vesicle invaginations from the plasma membrane in actin-coated vesicles [50] , potently restricted infection ( S6 Fig ) . In some scenarios , the maintenance of Arf6 in a GDP bound state may serve in a proviral , or promicrobial , manner . Recently , several Arf GAP domain-containing molecules , including ADAP1 , were shown to directly facilitate the uptake of Salmonella into host cell via their effects on the actin cytoskeleton [51] . Our findings presented here implicate ADAP2 as an ISG specifically induced to alter host cell endocytic and intracellular trafficking pathways to restrict viral entry . We show that the expression of ADAP2 dramatically induces macropinocytosis via an Arf6 GAP-dependent pathway , which correlates to alterations in the uptake of transferrin and in an inhibition of DENV and VSV entry and/or intracellular trafficking .
U2OS , 293T , Vero , 2fTGH ( STAT1 wild-type ) and U3A ( STAT1 mutant ) fibrosarcoma cells ( described previously [52] ) were grown in DMEM-H supplemented by 10% FBS and penicillin/streptomycin . HeLa ( CCL-2 ) cells were grown in MEM supplemented by 5% FBS and penicillin/streptomycin . HEK293T knockout MAVS were generated as follows: cells were plated at a density of 2×104 cells per well in a 96-well plate . The next day , CRISPR plasmids were transfected using GeneJuice transfection reagent ( Merk Millipore ) according to the manufacturer’s protocol . pRZ-mCherry-Cas9 and pLenti-gRNA constructs were transfected at a ratio of 3:1 ( i . e . 150 ng: 50 ng ) . Critical exons of MAVS were targeted using a gRNA construct ( sequence available by request ) . Subsequently , limiting dilution cloning was performed and after 10 days , growing monoclones were selected by bright field microscopy and positive clones trypsinized and expanded in two separate wells . One well was used to recover gDNA as previously described ( Ablasser et al . , 2013 ) and subsequently the target region of interest was amplified in a two-step PCR and subjected to deep sequencing . Knockout cell clones were identified as cell clones harboring all-allelic frame shift mutations using OutKnocker ( Schmid-Burgk et al . , 2014 ) . Genotype of the respective knockout cell line is available upon request . Experiments were performed with DENV-2 ( 16681 ) obtained from BEI Resources and expanded in C6/36 mosquito midgut cells as described previously [53] and titer was determined in Vero cells by a fluorescent foci forming unit ( FFU ) assay , as previously described [54] . Vesicular stomatitis virus ( VSV ) expressing GFP and coxsackievirus B3-RD ( CVB3-RD ) have been described previously [55] . Sendai virus ( SeV ) was purchased from Charles River Laboratories . Experiments measuring productive virus infection were performed approximately 48h post-transfection , at which time cells were infected with DENV-2 at a multiplicity of infection ( MOI ) of 1 FFU/cell for 24h , VSV and CVB3-RD at MOIs of 0 . 2 PFU/cell for 8h , SeV at 100 hemagglutination units ( HAU ) /mL , unless otherwise stated . ADAP2-V5 and GFP-ADAP2 were constructed by amplification of human ADAP2 cDNA ( clone Id: 5214358 , Thermo Scientific ) and cloning into pcDNA 3 . 1/V5-His TOPO TA Expression Kit or NT-GFP Fusion TOPO TA according to the manufacturer’s protocol ( Invitrogen ) . DsRed-fused ADAP2 was constructed by amplification of ADAP2 cDNA followed by insertion into the XhoI and EcoRI sites of pDsRed2-C1 ( Clontech ) . GFP-ADAP2-ΔArfGAP and GFP-ADAP2-ΔPH2 were constructed by amplification of ADAP2 cDNA with primers beginning at residue 132 or ending at residue 254 , respectively , followed by cloning into NT-GFP Fusion TOPO TA according to the manufacturer’s protocol ( Invitrogen ) . Mutagenesis was performed using Quikchange ( Stratagene ) according to the manufacturer’s protocol . RFP-LAMP1 ( plasmid #1817 ) , RFP-EEA1 ( plasmid #42635 ) , GFP-Rab8a ( plasmid #24898 ) , GFP-Rab8a[Q67L] ( plasmid #24900 ) , and pcDNA-HA-Arf6 ( plasmid # 10834 ) were obtained from Addgene . EGFP-tagged Rab34 and Rab7 constructs have been described previously [56] . YFP-Actin was kindly provided by Jeffrey Bergelson , Children’s Hospital of Philadelphia . Control ( scrambled ) siRNA and siRNA targeting ADAP2 ( 5’-GGACUGGUUCAAUGCCCUC-3’ ) were purchased from Sigma . Transfection of U2OS and 293T cells with plasmids was performed using X-tremegene 9 DNA ( Roche ) or X-tremegene HP DNA ( Roche ) transfection reagents according to the manufacturer’s protocol . SiRNAs were transfected using Dharmfect-1 according to the manufacturer’s protocol . Cells were infected and/or fixed 48h post-transfection . Rabbit anti-MAVS antibody was obtained from Bethyl Laboratories . Mouse anti-HA antibody , mouse anti-V5 , goat anti-EEA1 and rabbit anti-GAPDH HRP-conjugated antibodies were purchase from Santa Cruz Biotechnology . Mouse anti-VSVG ( P5D4 ) and mouse anti-DENV ( clone D3-2H2-9-21 ) antibodies were purchased from Santa Cruz Biotechnology and Millipore , respectively . Alexa Fluor 488 or 594 phalloidin and Alexa fluor-conjugated secondary antibodies were purchased from Invitrogen . We used high-throughput microarray analysis , performed as we previously described [57] , to screen for transcriptional changes in control ( 2fTGH ) vs . STAT1 signaling deficient ( U3A ) HT1080 cells , both treated with 100U of purified IFNβ ( PBL ) for 24hrs . In parallel , mock-treated 2fTGH and U3A were also included and were used to identify differentially expressed genes in IFNβ-treated cells . Briefly , the quality of all RNA samples was confirmed using an Agilent 2100 Bioanalyzer ( Agilent Technologies , Santa Clara , CA ) to ensure RNA integrity and quality . mRNA labeling was performed using a One-Color Low Input Quick Amp Labeling Kit ( Agilent ) and prepared for hybridization on SurePrint G3 Human Gene Expression 8x60K slides using the Gene Expression Hybridization Kit ( both from Agilent ) . Slides were scanned using Agilent’s SureScan Microarray Scanner System , and data extracted using Agilent’s Feature Extraction Software ( version 11 . 0 . 1 . 1 ) . Microarray data were normalized using the cyclic loess normalization method [58] . The R package Limma ( Linear Models for Microarray Data ) , which implements a moderated t-test , was used to identify differentially expressed mRNAs between mock- and IFNβ -treated samples [59] . Storey's q-value method [60] , as implemented in R package q-value , was used to calculate the adjusted p-values in order to control the false discovery rate . U2OS cells were transfected with indicated plasmids in glass-bottom 35mm dishes ( MatTek ) . Approximately 48 h following transfection , plates were placed into a 37°C , CO2-controlled incubator positioned over a motorized inverted microscope to allow for long-term time-lapse imaging ( VivaView FL; Olympus ) , and images captured every 10–15 min for ~24h , unless otherwise stated . In all experiments , cells cultured in 8-well chamber slides ( LabTek ) were washed and fixed with 4% paraformaldehyde followed by permeabilization with 0 . 1% Triton X-100 in PBS . Cells were incubated with the indicated primary antibodies for 1 hr at room temperature , washed , and then incubated with secondary antibodies for 30 min at room temperature , washed , and mounted with Vectashield ( Vector Laboratories ) containing 4’ , 6-diamidino-2-phenylindole ( DAPI ) . Images were captured using a FV1000 confocal laser scanning microscope ( Olympus ) , analyzed using Image J/Fiji ( NIH ) or Imaris ( Bitplane ) , and contrasted and merged using Photoshop ( Adobe ) . For three-dimensional analysis , xy or yz series stacks were acquired at ~0 . 5 μm intervals through the total thickness of the cell monolayer ( ~10 μm ) . Single particle tracking was performed using the MTrackJ plugin in Image J/Fiji . For imaging quantification , >20 individual organelles from at least three unique fields were measured using Imaris . For measurements of colocalization , >20 individual cells from at least three experiments were used to calculate Pearson’s colocalization coefficients using the Coloc2 plugin in ImageJ/Fiji . For EIPA treatment studies , U2OS cells were transfected with DsRed-ADAP2 for 48h , then treated with EIPA ( 102 μM , from Sigma ) for 60 min in complete medium , followed by fixation/permeabilization as described above . For studies related to viral entry , U2OS cells were transfected with vector or ADAP2 plasmids , as indicated , for ~48hrs . At this time , virus ( ~35 FFU/cell for DENV and ~100 PFU/cell for VSV ) was preadsorbed to cells for 60mins at 16°C . Following this incubation and a brief washing to remove unbound virus , viral entry was initiated by shifting the temperature to 37°C . At the indicated times ( ~30min for DENV and VSV ) , cells were fixed in 4% PFA and immunostained as described above . Total cellular RNA was extracted using TRI reagent ( MRC ) or a GenElute total RNA miniprep kit ( Sigma ) according to the manufacturer’s protocol . RNA samples were treated with RNase-free DNase ( Qiagen or Sigma ) prior to cDNA synthesis . Total RNA ( 1 μg ) was reverse transcribed by using iScript cDNA synthesis kit ( Bio-Rad ) . RT-qPCR was performed using iQ SYBR green supermix ( Bio-Rad ) in an Applied Biosystems StepOnePlus real-time PCR machine . Gene expression was calculated using the 2-△△CT method[61] , normalized to actin . QuantiTect primers against ADAP2 , IFI44L , DENV , VSV , were purchased from Sigma . Primer sequences were as follows: ADAP2 ( 5’-AAGCTGTCATCAGCATTAAG-3’ and 5’-ACTATCTCCTTCCCACTTTC-3’ ) ; IFI44L ( 5’-ACTAAAGTGGATGATTGCAG-3’ and 5’-TGCAGAGAGGATGAGAATATC-3’ ) ; DENV ( 5’-AGTTGTTAGTCTACGTGGACCGA-3’ and 5’-CGCGTTTCAGCATATTGAAAG-3’ ) . Actin , VSV , CVB , ISG56 , ISG60 and SeV primer sequences have been described [36 , 62] . U2OS cells were transfected as described above . Approximately 48h following transfection , cells were washed with PBS and incubated with complete medium containing 70 , 000-MW dextran conjugated to Oregon green 488 ( 0 . 1 mg/ml; Invitrogen ) or 10 , 000-MW dextran conjugated to Alexa Fluor 594 , as indicated , for 45 min followed by fixation/permeabilization as described above . For generation of NR-DENV , C6/36 cells were infected with DENV-2 at 33°C for 1 h in FBS-free DMEM ( DMEM-0 ) then incubated with DMEM containing 2% FBS ( DMEM-2 ) containing 100 μg neutral red dye ( Sigma ) for 5 days in the dark . Following this incubation , the medium was harvested , cleared from cellular debris by low-speed centrifugation , aliquoted , and stored at −80°C . NR-DENV titers were measured in either dark or light-exposed conditions in Vero cells using a foci forming unit assay as described [54] . All experiments with NR-DENV were performed under semidark conditions , unless otherwise stated . For experiments measuring NR-DENV infection , 293T cells were transfected with the indicated plasmids and ~ 48h post-transfection , cells were infected with NR-DENV ( MOI = 5 ) for 2 h in the dark . At this time , cells were illuminated on a light box for 20 min . In parallel , monolayers were maintained in the dark to control for effects unrelated to entry or illuminated at 0hr post-infection to verify the light sensitivity of NR-DENV . Cells were then washed , collected by manual pipetting , and then transferred onto naive 293T cells . Cells were infected for approximately 48h , washed and infection levels assessed by RT-qPCR . Cells were grown in 24-well plates and lysed in RIPA buffer [50 mM Tris-HCl ( pH 7 . 4 ) ; 1% NP-40; 0 . 25% sodium deoxycholate; 150 mM NaCl; 1 mM EDTA; 1 mM phenylmethanesulfonyl fluoride; 1 mg/ml aprotinin , leupeptin , and pepstatin; 1 mM sodium orthovanadate] . Lysates were sonicated and insoluble material was cleared by centrifugation . Protein concentration of lysates was determined by BCA protein assay ( Thermo Scientific ) . Lysates containing equal amounts of protein were loaded onto 4 to 20% Tris-HCl gels ( Bio-Rad ) and transferred to nitrocellulose membranes . Membranes were blocked in 5% nonfat dry milk , probed with the indicated antibodies , and developed with horseradish peroxidase-conjugated secondary antibodies ( Santa Cruz Biotechnology ) , and SuperSignal West Pico or Dura chemiluminescent substrates ( Pierce Bio-technology ) . Data are presented as mean ± SD unless otherwise stated , and were analyzed with Prism software ( Graphpad ) by two-tailed unpaired Student’s t-test . A p value <0 . 05 was considered significant . | The induction of antiviral innate immune signaling is a primary defense strategy employed by host cells to restrict virus infections . This system is triggered by the presence of ‘non-self’ components such as viral nucleic acids and culminates in the induction of type I interferons ( IFNs ) . Type I IFNs themselves possess no direct antiviral activity and instead exert their potent antiviral effects via the induction of hundreds of interferon-stimulated genes ( ISGs ) that directly antagonize viruses at a variety of steps in their infectious cycles . Here we identify ArfGAP with dual pleckstrin homology ( PH ) domains 2 ( ADAP2 ) as an ISG whose expression restricts the replication of dengue virus ( DENV ) , which infects as many as 400 million people worldwide annually , at the stage of virion entry and/or trafficking . Our study thus not only identifies ADAP2 as a previously uncharacterized ISG , but also points to its role in the inhibition of DENV replication at the earliest stages of the DENV life cycle . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | ADAP2 Is an Interferon Stimulated Gene That Restricts RNA Virus Entry |
Cohesin is a chromatin-associated protein complex that mediates sister chromatid cohesion by connecting replicated DNA molecules . Cohesin also has important roles in gene regulation , but the mechanistic basis of this function is poorly understood . In mammalian genomes , cohesin co-localizes with CCCTC binding factor ( CTCF ) , a zinc finger protein implicated in multiple gene regulatory events . At the imprinted IGF2-H19 locus , CTCF plays an important role in organizing allele-specific higher-order chromatin conformation and functions as an enhancer blocking transcriptional insulator . Here we have used chromosome conformation capture ( 3C ) assays and RNAi–mediated depletion of cohesin to address whether cohesin affects higher order chromatin conformation at the IGF2-H19 locus in human cells . Our data show that cohesin has a critical role in maintaining CTCF–mediated chromatin conformation at the locus and that disruption of this conformation coincides with changes in IGF2 expression . We show that the cohesin-dependent , higher-order chromatin conformation of the locus exists in both G1 and G2 phases of the cell cycle and is therefore independent of cohesin's function in sister chromatid cohesion . We propose that cohesin can mediate interactions between DNA molecules in cis to insulate genes through the formation of chromatin loops , analogous to the cohesin mediated interaction with sister chromatids in trans to establish cohesion .
Cohesin is an evolutionarily conserved protein complex composed of the core subunits , SMC1 , SMC3 , SCC1/RAD21 and SCC3/SA ( reviewed in [1] ) . It has been proposed that cohesin mediates sister chromatid cohesion by embracing replicated DNA molecules as a ring [2] . Cohesin also has important roles in gene regulation in yeast , animals and humans ( reviewed in [1] , [3] ) . This regulatory function also exists during G1 phase and in post-mitotic cells , indicating that cohesin affects gene expression independent of its role in cohesion [4]–[6] . Cohesin mediates gene regulation at least in part by interaction with insulator elements [6]–[11] . Insulators are chromatin boundaries that separate gene promoters from regulatory elements . The only known protein that directly binds insulators in mammalian cells is the multi-functional zinc finger protein CTCF ( CCCTC binding factor , reviewed in [12] ) . Several studies have recently identified co-localisation of CTCF and cohesin in mammalian genomes and have shown that CTCF is needed to recruit cohesin to these binding sites [6] , [9]–[11] . Remarkably , although CTCF can associate with its binding sites in the absence of cohesin , its enhancer blocking activity depends on cohesin [6] , [10] . It has , therefore , been speculated that CTCF may mediate transcriptional insulation by recruiting cohesin to particular sites in the genome [6] , but it remains unknown how cohesin controls gene regulation at these sites . The IGF2-H19 locus plays a role in the aetiology of embryonic growth disorders and in various cancers ( reviewed in [13] ) . A CTCF mediated insulator sequence plays a role in the reciprocal imprinting of IGF2 and H19 genes [14]–[16] . This insulator is located upstream of the H19 gene and is known as the imprinting control region ( ICR ) . It acquires methylation on the paternal allele during male germ cell development and is therefore also called the H19 differentially methylated region ( DMR ) or domain ( DMD ) . CTCF binding at the insulator prevents the IGF2 gene from accessing enhancers downstream of the H19 gene . This physical separation is thought to maintain the silence of the maternal IGF2 allele . On the methylated paternal allele CTCF is excluded from binding and the IGF2 promoters can interact with the enhancers [14] , [15] . In mice it has been demonstrated that higher order chromatin conformation at this locus differs between the maternal and paternal alleles and that CTCF binding is essential for the formation of chromatin loops on the maternal allele [17]–[22] . In addition to the ICR and the downstream enhancers , several additional regulatory regions have been described surrounding the locus . At the 5′end of IGF2 there is a differentially methylated region , DMR0 , which in human has variable methylation in somatic tissues and is hypomethylated in cancers [23]–[25] . In the intervening sequence between IGF2 and H19 there is a Centrally Conserved DNase I hypersensitive domain ( CCD ) , which in mice has tissue specific enhancer functions [26]–[30] . In humans the functions of the CCD and DMR0 are unknown . Because CTCF is needed for the recruitment of cohesin to insulator sequences , we speculated that cohesin , with its unique capability of holding DNA strands together , is required for the formation and stabilisation of CTCF-dependent chromatin loops . To test this hypothesis we undertook quantitative Chromatin Conformation Capture ( q3C ) analysis [31] of the human IGF2-H19 locus ( Figure 1 ) and examined CTCF-dependent chromatin loops after cohesin depletion by RNAi . Our results indicate that cohesin has an important role in long-range interactions between CTCF sites , whereas CTCF independent chromatin associations do not require the presence of cohesin . Cohesin may , therefore , contribute to gene regulation at CTCF sites by mediating the formation of chromatin loops .
To predict higher order interactions that may be directly mediated by CTCF and cohesin at this locus in HB2 cells , we performed locus-wide ChIP experiments with antibodies specific for CTCF and SMC3 ( this subunit of cohesin was analysed due to the availability of validated ChIP grade antibodies [6] ) . We found that CTCF and cohesin co-localise at the ICR , at CTCF sites immediately adjacent to the DMR0 , at CTCF sites in the CCD region [26] , [27] , [30] and also with CTCF sites downstream of the enhancers ( Figure 2A and 2B ) , consistent with previous data from other cell lines [6] , [35] . We refer to the CTCF sites adjacent to the DMR0 as “CTCF AD” and to those downstream of the enhancers as “CTCF DS” ( Figure 1 ) . In contrast to the above mentioned sites , CTCF and cohesin were not enriched at the DMR0 , IGF2 promoters or the enhancers downstream of H19 ( Figure 2A and 2B ) . SNP sequencing of ChIP-PCR samples indicated that CTCF and cohesin bind at all sites to both alleles , with exception of the ICR , where monoallelic binding was observed on the unmethylated allele ( Figure 2C and 2E ) . The CTCF sites adjacent to the DMR0 and the CCD are not CpG rich and were unmethylated ( data not shown ) . We determined the chromatin conformation at the human IGF2-H19 locus through extensive q3C analyses with a BamHI restriction enzyme , using primers within the ICR , enhancer , and two other CTCF sites ( CTCF AD/DMR0 and CTCF DS ) flanking the locus as anchors . BamHI cuts the locus frequently , but there are no restriction sites within the CTCF AD . The nearest site to the CTCF AD is on the edge of the DMR0 region ( restriction site a ) . For this reason we have regarded CTCF AD/DMR0 as a single 3C element in our analysis . The resulting data represent an average of association frequencies on both parental alleles across the whole locus . Random ligations are expected to decrease exponentially the further a restriction site is away from the anchor , while specific associations occur as “spikes” above the random ligation curve [36] . However , because the resolution of 3C is limited , multiple adjacent restriction sites within a 5 Kb stretch of DNA will associate with similar frequencies with a distant restriction site used as an anchor . The 3C signals therefore indicate proximities to interactions , rather than pinpoint the exact sequences involved in the interactions . When a primer within the ICR was used as an anchor ( primer k ) , weak but specific associations were detected with the CTCF AD/DMR0 region ( restriction site a , Figure 3B and 3C ) . When the enhancer was used as an anchor ( primer m ) we were able to detect associations with a restriction site within the CTCF AD/DMR0 ( restriction site a ) as well as at restriction sites near the P2 and P3 promoters ( sites b2 , c1 and d , Figure 3D and 3E ) , despite low levels of IGF2 expression in these cells . However , when an anchor primer was placed within the CTCF AD/DMR0 region ( primer b1 ) , we found strong association frequencies with the CCD ( restriction site h ) and the CTCF DS ( restriction site q ) , but not with the ICR or the enhancer ( Figure 3F ) . Because the DMR0 does not bind CTCF , it is possible that the associations between the CTCF AD/DMR0 and the CCD are caused by CTCF binding at the CTCF AD . To confirm that the CTCF AD/DMR0 region interacts with the distant CTCF DS sites we also used a primer within the CTCF DS ( primer q ) as an anchor . As expected , this revealed associations between CTCF DS and the CTCF AD/DMR0 region ( restriction sites a and b1 ) , and in addition showed associations with the CCD ( restriction site h ) ( Figure 3G ) . Locus-wide q3C analysis was also performed using BglII as a restriction enzyme which cuts in the CTCF AD site . These experiments revealed interactions between the CTCF AD anchor , the CCD and CTCF DS as well as ICR anchor and the CTCF AD ( Figure S1 and Figure S2 ) , similar to what we had obtained for BamHI experiments . These data indicate that the strongest associations that can be detected at the human IGF2-H19 locus in HB2 cells exist between the CTCF AD/DMR0 region and the CCD and CTCF DS sites , whereas a weak interaction may also exist between the ICR and CTCF AD/DMR0 . The interaction between the CTCF AD/DMR0 and the ICR could be weaker than the CTCF AD/DMR0 and the CCD because the ICR is a monoallelic CTCF/cohesin binding site , and/or because our ChIP experiments indicated that less CTCF and cohesin are bound to the ICR than to the other sites ( Figure 2A and 2B ) . Alternatively , we cannot exclude the possibility that the 3C interactions that we detected between the ICR and CTCF AD/DMR0 were caused indirectly by the strong associations between CTCF AD/DMR0 and the CCD and the CTCF DS sites , which are located to the left and the right of the ICR . To examine which of the detected chromatin associations is allele specific we combined the 3C assays with SNP analysis of ligated products . Using polymorphisms in two separate restriction fragments , we detected associations between the enhancer and the IGF2 promoter region predominantly on one allele ( Figure 4A and 4B ) , which presumably represents the paternally derived allele . These promoter-enhancer interactions suggest that , despite the low levels of IGF2 transcription , the chromatin conformation in HB2 cells is favourable for monoallelic expression . Associations between the CTCF AD/DMR0 region and the CCD could be detected on both alleles , consistent with our finding that CTCF and cohesin bind to CTCF AD sites and the CCD biallelically ( Figure 4C ) . Monoallelic interactions were detected on the CTCF binding allele ( presumably the maternal one ) between the ICR and the CTCF DS ( Figure 4D ) . Unexpectedly , we also detected biallelic associations between the ICR and the CTCF AD/DMR0 region ( Figure 4E ) , despite the fact that CTCF binds the ICR predominantly on one allele . This result suggests that ICR- CTCFAD/DMR0 interactions are indirect and mediated through interactions between the CTCF AD/DMR0 with other CTCF sites near the ICR such as the CTCF DS . It is also conceivable that elements within the DMR0 and the adjacent CTCF sites interact separately with the methylated and the unmethylated ICR . However , most regulatory elements are quite close to the ICR and it is , therefore , also possible that the paternal promoter-enhancer interactions distort the conformation of the loop in a manner such that the paternal ICR may also be in close proximity to IGF2 . Because our ChIP experiments revealed the strongest CTCF and cohesin signals at the CTCF AD , CCD and CTCF DS sites ( Figure 2A and 2B ) , and our q3C assays had identified the strongest interactions between these sites , we hypothesised that these interactions could be mediated by cohesin . First , we tested two predictions that are made by this hypothesis . Cohesin associates with chromatin throughout interphase and regulates transcription at the IGF2-H19 locus both in G1 and G2 phase [6] . If cohesin affects gene regulation by enabling the formation of chromatin loops one would , therefore , predict that these loops are also present in G1 and G2 phase . To test this we used 3C assays to compare chromatin conformation between cells that were synchronised by double thymidine arrest-release either in G1 or G2 phase . Figure 3 shows that synchronisation of cells in G1 and G2 phase did not change the overall 3C profiles using the CTCF AD/DMR0 , ICR or enhancers as anchors . An exception was observed for the association between the ICR anchor and the CTCF DS which did not stand out as a “peak” in G1 cells , but could be detected as a “shoulder” in the 3C profile of the G2 cells . We presently do not know if this difference is of physiological relevance . Since most interactions did not change between G1 and G2 phase , our results are consistent with the possibility that cohesin mediates chromatin loop formation throughout interphase . Importantly , these data also suggest that cohesin's role in chromatin conformation would have to be independent of its function in sister chromatid cohesion , which exists in G2 but not G1 phase . Another prediction made by the hypothesis that cohesin mediates 3C interactions between CTCF sites at the IGF2-H19 locus is that cohesin should be present in the corresponding chromatin loops . Since our cohesin ChIP experiments ( Figure 2A and 2B ) did not distinguish between binding of cohesin to DNA molecules which were folded into loops and those that were not , we enriched the 3C digested templates for CTCF or cohesin bound chromatin before ligating for 3C analysis ( ChIP-loop ) . We found that both SMC3 and CTCF antibodies can immunoprecipitate 3C products representing associations between the ICR ( j restriction site ) and the CTCF AD/DMR0 region ( b1 restriction site ) ( Figure 5 ) . These results indicate that cohesin is present in the corresponding chromatin loops . Importantly , 3C products representing enhancer-promoter associations could not be detected . These observations indicate that enhancer-promoter interactions cannot directly be mediated by cohesin , whereas the data are consistent with the possibility that cohesin mediates chromatin association between CTCF sites . To test directly if cohesin is functionally required for the formation or maintenance of chromatin loops at the IGF2-H19 locus , we depleted the cohesin subunit SCC1 by RNAi and thus rendered the cohesin complex non-functional . Since cohesin depleted cells delay progression through mitosis , and because it is unknown if chromatin loops are maintained during mitosis , we synchronised SCC1 depleted cells by double thymidine treatment and harvested cells in G1 and G2 phases ( Figure 6A and 6B ) . We confirmed that CTCF was still bound to the ICR , and the CCD after cohesin depletion in HB2 cells ( results not shown ) , as was previously demonstrated in another cell line [6] . The effect of depletion of SCC1 on IGF2 and H19 expression was an activation of IGF2 transcription , but no significant difference in H19 expression ( Figure 6D ) . Average DNA methylation levels at the IGF2 and H19 DMRs did not change significantly ( Figure 6E and 6F ) . In SCC1 depleted cells that were enriched in G2 phase we found a significant reduction in the association frequency between all chromatin interactions between CTCF binding sites , with the exception of the interaction between the CTCF AD/DMR0 region and the very distant CTCF DS sites ( Figure 7B , 7C , 7G , and 7H and Table S1 ) . Similar results were obtained using an ICR anchor primer in another restriction site within the ICR ( Primer j , Figure S3 ) . Using BglII as a restriction enzyme we were also able to confirm that there is a 30% reduction between the CTCF AD and CCD interactions during G2 phase ( Figure S2D ) . Importantly , however , associations with the enhancer anchor ( primer m ) and the IGF2 promoters ( restriction sites b2 , c1 ) were not significantly reduced ( Figure 7D and 7E ) . These observations indicate that 3C interactions between CTCF sites are dependent on cohesin , whereas the ability of the enhancer to associate with the IGF2 promoters is not dependent on cohesin . Importantly , when we sequenced the PCR products from the enhancer anchor , we found that the monoallelic associations found in the controls became biallelic after cohesin depletion ( Figure 7F ) . SCC1 depleted cells harvested in G1 phase yielded less chromatin for analysis than that in G2 phase and locus wide comparisons could not be done . However we were able to show that in G1 , associations between the ICR ( j primer ) and the CTCF AD/DMR0 regions ( restriction sites a- c1 ) were also reduced ( Figure S4 ) . Together , these results indicate that depletion of cohesin predominantly affects CTCF mediated looping interactions . Importantly , our observation that cohesin depletion also affects chromatin structure in G1 phase , where no sister chromatid cohesion exists , further supports the notion that cohesin's role in chromatin looping is independent of its cohesion function . The effect of decreased CTCF and cohesin mediated 3C associations had little or no impact on imprinted expression of H19 which remained mono-allelic ( Figure 8B ) , presumably due to the methylation being maintained at the ICR ( Figure 6E and 6F ) . Expression of IGF2 was biallelic after cohesin depletion as determined by SNP analysis and RNA FISH ( Figure 8A and 8C ) . However , IGF2 expression levels were so low prior to cohesin depletion that was impossible to tell whether the basal transcription was monoallelic or biallelic .
It is now well established that cohesin complexes do not only function in sister chromatid cohesion but also have important roles in gene regulation , both in proliferating and post-mitotic cells ( reviewed in [3] , [37] ) . However , it remains largely unknown how cohesin mediates these effects . Our results suggest that cohesin contributes to gene regulation by mediating the formation of higher order chromatin conformation , at least at the imprinted IGF2-H19 locus . Recent studies of the apolipoprotein cluster [38] and of the interferon γ locus [39] have revealed that cohesin also has roles in long-range chromatin interactions at these loci . It is , therefore , possible that cohesin has a widespread role in the formation of chromatin loops in mammalian genomes and regulates gene expression through this mechanism at numerous sites . In our experiments 3C interactions were reduced but not abolished after cohesin depletion . It is possible that the remaining chromatin interactions were caused by residual amounts of cohesin , which is difficult to deplete completely by RNAi . However , we can not exclude the possibility that other proteins maintain the chromatin loops in the absence of cohesin . Our first systematic q3C analysis of the human IGF2-H19 locus has brought to light an unexpected complexity of chromatin interactions . We found evidence for IGF2 promoter-enhancer interactions on the putative paternal allele [17] , [18] as well as ICR interactions with CTCF sites at the 5′ end of the IGF2 gene ( CTCF AD sites ) . Previous studies in mice have identified allele specific interactions of the ICR and the enhancers , but CTCF sites other than the ICR have not yet been analysed at the mouse locus . By extending our analysis to a wider number of CTCF sites we found a previously unknown association of the ICR with a CTCF DS site on the presumed maternal allele in the human cell line , as well as biallelic CTCF mediated interactions involving the CCD site . The CCD region in mice has tissue specific silencer or enhancer activities which are independent of imprinting [26] , [27] , [30] . Genome-wide CTCF ChIP-sequencing data in adult mouse livers indicate that this region also binds CTCF ( D . Odom and D . Schmidt , pers . communication ) , suggesting that the CCD may function as a boundary or insulator with regard to its silencer function . It is thus possible that cohesin is also required for the insulator activity of the CCD . Our data are consistent with the possibility that multiple CTCF-cohesin mediated loops come together in a chromatin “hub” as depicted in Figure 9 . At this hub CTCF and cohesin might bring various regulatory elements into close proximity to enable interactions between distant elements , either simultaneously as is drawn in our model , or possibly in a sequential order . Cohesin may stabilise the CTCF mediated interactions . Cross-linked chromatin enables us to study a snapshot of interactions at any given time , but it is likely that these interactions are dynamic with some occurring more rapidly than others . Transcription of IGF2 and H19 is developmentally down regulated in most adult tissues but reactivated in various cancers ( reviewed in [13] ) . We chose a normal epithelial breast cell line to study the conformation of the adult IGF2-H19 locus . After the disruption of CTCF mediated chromatin conformation by cohesin depletion IGF2 expression was reactivated in these cells . Moreover , substantial biallelic expression of IGF2 was observed and enhancer-promoter associations changed from mono- to biallelic . Interestingly , biallelic IGF2 expression was not accompanied by hypermethylation at the ICR , suggesting that depletion of cohesin can uncouple the relationship between IGF2 expression and methylation at the H19 ICR . Methylation profiles of the IGF2-H19 locus in many cancers indicate that loss of IGF2 imprinting and methylation are often disconnected during neoplasia [40]–[46] . The roles of higher order chromatin structure and loss of imprinting in cancer are still largely unexplored . Defects in proper positioning of cohesin on DNA could therefore contribute to abnormal gene regulation in neoplastic cells . The finding that cohesin is required to stabilise higher order chromatin conformation raises the intriguing possibility that cohesin physically connects two DNA sequences on the same DNA molecule in cis , to form loop structures similar to how cohesin interacts with two DNA molecules in trans , to mediate sister chromatid cohesion . It is conceivable that cohesin forms chromatin loops by embracing two DNA strands at the base of a loop , similar to how cohesin has been proposed to mediate cohesion as a ring [2] . Alternatively , it is possible that cohesin complexes bound to two DNA sites can interact with each other , as has been suggested for CTCF molecules [47] . Our finding that cohesin depletion interferes with chromatin loop formation , although cohesin depletion does not abrogate CTCF binding [6] , supports our hypothesis that one of CTCF's main roles as a transcriptional insulator may be to recruit cohesin to insulator sequences . Hypomorphic mutations in the cohesin subunits SMC1 and SMC3 and in the cohesin loading factor NIPBL have been identified as the molecular cause of Cornelia de Lange syndrome , a rare human developmental disorder [48]–[50] . The results in this study show a modest reduction in looping interactions after cohesin depletion which suggest that cohesin is a stabilising factor in chromatin looping . It will therefore be important to test if these mutations affect higher order chromatin structure at specific loci . Although such defects may be very subtle , they could at some loci cause defects in gene regulation during development . Another important goal for the future will be to determine at the genomic level which CTCF-cohesin sites can interact with each other , if these interactions change during cell differentiation and how such changes might be specified .
Methylation analysis was by bisulphite- and pyrosequencing with primers as described previously [23] , [25] , [51] . Expression analysis was done by qPCR on reverse transcribed RNA . Primers for qPCR were IGF2 Fwd CTCACCTTCTTGGCCTTCG , IGF2 Rev GGAAACAGCACTCCTCAACG , H19 t Fwd GAGATTCAAAGCCTCCACGACT and H19 Rev GCGTAATGGAATGCTTGAAGG . B Actin was analysed using primers from a QuantiTect Assay ( Qiagen ) . Quantitation was done by extrapolation to standard curves for the pimers . Wilcoxon signed rank tests were done to compare paired RNAi and control samples when n≥3 . P≤0 . 05 was considered significant . Chromatin Immunoprecipitation ( ChIP ) was done as described previously [6] . Input and immunoprecipitated ( IP ) material was quantified by Picogreen ( Invitrogen ) , and real-time PCR with standard curves was performed . Values were corrected for DNA amount , and enrichment was calculated as IP over input . When comparing ChIP from cohesin depleted cells with control cells ( Figure 6C ) , the IP/Input was further normalised against a region where CTCF does not bind ( IGF2 exon 9 ) . ChIP primers are shown in Table S2 . Quantitative 3C was described previously [31] , [36] , [52] , [53] and performed with the following modifications: 5×106 of cells were cross-linked in 1% formaldehyde at 37°C for 10 minutes . After washing in PBS , the cells were lysed on ice in lysis buffer ( Tris-HCl ( pH = 8 ) 50 mM , SDS 1% , EDTA 10 mM ) for 10 minutes . Nuclei were recovered by centrifugation and resuspended in BamHI digestion buffer ( New England Biolabs ( NEB ) ) , supplemented with Triton-X100 to a final concentration of 1 . 8% and incubated 1 h at 37°C . 1 . 5×106 nuclei were digested overnight with 1000 U of BamHI ( NEB ) in a 300 µl reaction volume . Digestion efficiency at each BamHI restriction site within the locus was assessed by qPCR across each restriction site . The percentage of digestion was determined by comparing template amplification of digested and undigested fractions ( not religated ) after normalising to copy number as previously described [31] . All regions within the locus were digested equally efficiently . This step was an important quality control check , and if digestion was below 70% the chromatin was discarded . Ligation was carried out on 2 . 5 ng/µl digested chromatin in a 1 . 5 ml reaction volume of T4 ligase buffer containing 3200 U of T4 ligase ( NEB ) . A further overnight digestion step with 1000 U of EcoRI ( which cuts outside the hybrid religated products ) was incorporated prior to reversal of cross links , phenol chloroform purification and ethanol precipitation . This step is necessary to remove possible qPCR bias caused by size differences in the religated products . 3C PCR primers flanking restriction sites were designed to have similar melting temperatures , and the PCR efficiency of each primer combination was assessed on a PCR standard template . A stock of PCR standard template was prepared similar to that described previously [52] by amplification of 36 genomic regions across the IGF2-H19 locus on commercially obtained genomic DNA ( Becton Dickinson ( BD ) ) . These amplicons were column purified and quantified using Nanodrop UV spectroscopy . Equimolar amounts of amplicons were mixed , BamHI digested , re-ligated , phenol-chloroform extracted , ethanol precipitated , dissolved in H2O and stored at −20°C . Q-PCR was done with Sybr-green ( ABI Power SYBR ) on a 384 well real time machine ( 7900HT Fast Real time PCR system , Applied Biosystems ) . Quantitative determination of association frequencies was essentially done as described [31] . Copy number of 3C template was determined by qPCR amplification of a region between IGF2 and H19 which did not have BamHI restriction sites ( Chr11: 2057922–2057991 , Ensembl ) . Template copy number was used to ensure that the amount of 3C template was within the range of the standard curve for any given product . All interaction frequencies were normalised to the circularisation frequency of the i-fragment as internal digestion-ligation control . We verified our normalisation method by including a β-actin gene region ( Chr7: 5 , 326 , 283–5 , 357 , 206 ) that contained 3 BamHI restriction sites and compared the outcome of the 3C association frequencies across the locus when normalised to adjacent BamHI sites ( 2–3 B-actin ) , alternative BamHI sites ( 1–3 B-actin ) , or alternative internal sites within the IGF2-H19 locus ( data not shown ) . 3C primers and combinations are in Table S2 and S4 . Biological replicates for siRNA 3C experiments were done by splitting test and control cells each into 3 equal aliquots prior to synchronisation . After harvesting the cells and digesting the chromatin for 3C experiments , the replicate templates were evaluated for digestion efficiency and normalised for equal amounts before ligation . Prior to PCR amplification , the replicate templates' copy numbers were determined by qPCR as described above , and equal amounts ( copy number ) of DNA recovered after 3C for control and RNAi template was used for the 3C qPCR . Association frequencies were normalised to the circularisation frequency of the i-fragment as described above . Normalisation data for 3 biological replicates are shown in Figure S5 . The frequency of circularisation of the i-fragment is similar in controls and RNAi treated cells ( Figure S5A ) , confirming that RNAi treatment for cohesin depletion did not affect digestion or ligation efficiency . Using the circularisation frequency of the i-fragment to normalise an association frequency between the enhancer anchor ( primer m ) and a restriction site located between the enhancer and CTCF DS ( restriction site p ) , shows no significant differences between RNAi treated cells and controls ( Figure S5B ) . This is as expected , because there is no binding of CTCF or cohesin to the enhancer and its interaction with restriction site p is due to random ligation . Circularisation of the i-fragment was therefore a suitable internal normaliser . ChIP loop was performed as follows: 3 aliquotes of 5×106 cells were first fixed in formaldehyde and nuclei were prepared as for 3C . One aliquot was used for 3C and the remaining aliquots were briefly sonicated to produce chromatin fragments of 500 bp and then digested overnight with BamH1 as for 3C protocol . After digestion the nuclei were pre-cleared on agarose beads ( UPSTATE ) and immunoprecipitated with antibodies to cohesin ( anti-SMC3 [6] , [54] and CTCF ( UPSTATE ) as described in the ChIP protocol . After washing in ChIP washing buffer ( UPSTATE ) , the beads and antibody complexes were resuspended in 1 . 5 ml of 3C ligation mix , containing 3200 U of T4 ligase , overnight at 15°C . After ligation , the samples were purified as in the 3C protocol . After 40 cycles of PCR , bands were visualised on an agarose gel and compared to the band obtained with 3C . RNAi knockdown of cohesin was done as described previously [6] . To obtain cells enriched in G2 and G1 phase HB2 cells were synchronised by double thymidine block: addition of 3 mM thymidine for 16 h , removal of thymidine by washing with PBS and release of the cells from the block for 8 h , addition of 3 mM thymidine for another 16 h for the second block . The cells are released from the second block by washing with PBS and cells are harvested after 6 h for enrichment in G2 phase and after 14 hours for enrichment in G1 phase . The enrichment of the cells in the respective cell cycle phases was controlled by FACS . To obtain cells enriched in G1 and G2 phase and depleted of the cohesin subunit SCC1 the siRNA transfection was performed either 6 hours before starting the first thymidine block ( for G2 phase ) or 2 hours after releasing the cells from the first thymidine block ( for G1 phase ) . The siRNA oligos ( sense-GGUGAAAAUGGCAUUACGGtt and antisense CCGUAAUGCCAUUUUCACCtt , Ambion ) were annealed according to manufacturer's instruction and used at a final concentration of 75 nM . The siRNA transfection was performed using lipofectamine RNAiMAX ( Invitrogen ) . Two-way ANOVA with Bonferroni's post-test was performed using GraphPad Prism version 5 . 01 for Windows , GraphPad Software , San Diego California USA , www . graphpad . com . Normalised values of 3 biological replicate experiments for each ligation combination in a given anchor set was analysed by two-way ANOVA , with RNAi/control being one set of factors and restriction sites being the other set of factors . A Bonferrroni post post-test enabled comparison of multiple replicates at each restriction site . The Bonferroni correction lowers the P value considered significant to 0 . 05 divided by the number of comparisons . Thus in n rows of data with two columns ( Control and RNAi ) , the P value has to be less than 0 . 05/n , for any particular row in order to be considered significant with P<0 . 05 . This correction ensures that the 5% probability applies to the entire set of comparisons , and not separately to each individual comparison . To obtain a probe for RNA FISH two PCR products of 2000 bp and 600 bp ( primer sequences available upon request ) were generated from the last exon of the human IGF2 gene , mixed and labeled with dig-11-dUTP using the Biotin High Prime Kit ( Roche ) . For the DNA probe the human BAC RP11-650021 spanning the IGF2 gene ( chr . 11: pos . 2057305–2245714 ) was directly labeled with Alexa 594 by random priming . Cells on coverslips were fixed for 15 min with 4% formaldehyde , 5% acetic acid in PBS and stored after another PBS wash in 70% ethanol at 4°C . Denaturation and hybridisation of the slides and probes was done as described in [54] . The biotin labelled probe was detected by successive incubations with mouse-anti biotin ( DAKO , 1∶500 dilution ) and FITC-conjugated goat-anti-mouse antibodies ( Jackson ImmunoResearch Laboratories Inc . , 1∶500 dilution ) and after dehydration mounted using Vectashield with DAPI ( Vector Laboratories ) . The slides were analysed on a Leica DMRBE microscope equipped with a Hamatsu CCD ( C4880 ) camera with a 100X objective . Adobe Photoshop was used to colour the images and generate the overlay figures . | Recent work has shown that cohesin , a protein best known for its role in holding sister chromatids together , and CTCF , a protein implicated in the formation of chromatin loops , localize to the same regions of DNA in mammalian genomes . This observation raised the intriguing possibility that cohesin might facilitate the role of CTCF in structuring chromatin . CTCF is well known for its role in regulating genomic imprinting at the IGF2-H19 gene locus . Imprinted genes are widely studied due to their roles in fetal growth and cancer and have the unusual property of expressing only one parental copy of the gene . CTCF is thought to regulate imprinting of IGF2 and H19 by enabling DNA to form loops that separate the genes into silent or active domains . In this paper we describe , for the first time , the looping structure of the human IGF2-H19 locus and show that cohesin stabilises CTCF–mediated DNA loops . Depletion of cohesin leads to disruption of long-range chromatin interactions and changes expression levels of the IGF2 gene . This work adds a new level of understanding of how cohesin can play a role in gene expression . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"genetics",
"and",
"genomics/animal",
"genetics",
"genetics",
"and",
"genomics/chromosome",
"biology",
"molecular",
"biology/dna",
"methylation",
"genetics",
"and",
"genomics/epigenetics",
"genetics",
"and",
"genomics/cancer",
"genetics",
"molecular",
"biology/chromatin",
"structure"
] | 2009 | Cohesin Is Required for Higher-Order Chromatin Conformation at the Imprinted IGF2-H19 Locus |
Parvoviruses exploit transferrin receptor type-1 ( TfR ) for cellular entry in carnivores , and specific interactions are key to control of host range . We show that several key mutations acquired by TfR during the evolution of Caniforms ( dogs and related species ) modified the interactions with parvovirus capsids by reducing the level of binding . These data , along with signatures of positive selection in the TFRC gene , are consistent with an evolutionary arms race between the TfR of the Caniform clade and parvoviruses . As well as the modifications of amino acid sequence which modify binding , we found that a glycosylation site mutation in the TfR of dogs which provided resistance to the carnivore parvoviruses which were in circulation prior to about 1975 predates the speciation of coyotes and dogs . Because the closely-related black-backed jackal has a TfR similar to their common ancestor and lacks the glycosylation site , reconstructing this mutation into the jackal TfR shows the potency of that site in blocking binding and infection and explains the resistance of dogs until recent times . This alters our understanding of this well-known example of viral emergence by indicating that canine parvovirus emergence likely resulted from the re-adaptation of a parvovirus to the resistant receptor of a former host .
The interactions between viral pathogens and their hosts present a longstanding evolutionary challenge for both participants . Viruses are continuously selected for improved replication and spread in host populations , while hosts are reciprocally selected for increased resistance to infection . Thus , the viruses that exist today have been shaped by a sustained interplay with hosts over long periods of evolutionary time [1] . Much attention has been focused on the evolution of viruses , but less is known about the corresponding variation and selection of relevant host genes . However , it is clear that pathogen-driven selective pressures can also drive genetic change in the host genes that control susceptibility and disease progression . The analysis of these evolutionary interplays helps elucidate the factors that control viral emergence , defined here as the establishment of an existing virus in a novel host species . Canine parvovirus ( CPV ) arose in the mid-1970s , and spread world-wide in 1978 as the cause of a new disease pandemic in dog , and that virus was clearly a variant of feline panleukopenia virus ( FPV ) . CPV has continued to circulate among dogs throughout the world , causing significant clinical disease [2] . Parvoviruses are single-stranded DNA viruses , and the CPV- and FPV-like viruses are ubiquitous in nature , infecting most animals among the order Carnivora [3] . Viruses of the family Parvoviridae have circulated widely amongst many animal hosts for millions of years , as was revealed through the identification of ancient parvovirus genomes and genome fragments captured by vertebrate genomes tens of millions of years ago [4]–[6] . A group of parvoviruses closely related to feline panleukopenia virus ( FPV ) infects many hosts among the order Carnivora ( carnivores ) , but domestic dogs , wolves , coyotes and some related carnivores resisted infection by those viruses until the emergence of CPV in 1978 [7] . The emergence of CPV in dogs was associated with the virus acquiring the ability to bind the canine transferrin receptor type-1 ( TfR ) [8] . While parvoviruses have certainly been evolving and changing over evolutionary time , it seems that they could also be providing a selective pressure that would conversely shape the evolution of key host genes that modulate their success , such as the TFRC gene that encodes TfR . Indeed , there is clear evidence of selection on host receptors in other viral-host systems , including MHC [9]; CD4 [10] , and Toll-Like Receptors [11] , and we wondered if the same could be true for TfR . This is important to address because viral and host controls of infection sit at the heart of our understanding of how novel viruses can emerge . The TfR is a dimeric type II membrane protein , where each monomer is comprised of carboxypeptidase-like , helical , and apical domains , as well as stalk , transmembrane and cytoplasmic sequences . The normal function of TfR is to bind iron-loaded transferrin ( Tf ) via the carboxypeptidase-like and helical domains [12] , [13] and mediate clathrin-mediated endocytosis . The TfR also binds to the hemochromatosis ( HFE ) protein which competes with the binding of Tf to regulate the uptake of iron from the intestine [14] . In our previous studies of CPV and FPV binding to the feline and canine TfR , apical domain residues were seen to be critical for controlling parvovirus binding [15] , [16] . One key mutation in this domain , present in dog TfR , introduces a novel N-linked glycosylation site at Asn384 of that TfR ( equivalent to feline TfR Lys383 ) . This glycan , together with other sequence determinants in the apical domain , collectively create the block to FPV binding observed for dog cells , with the Asn384 mutation having the greatest effect [15] . As such , changing only residue 384 in the canine TfR from Asn to the feline encoded Lys allows efficient FPV binding , while the replacement in the equivalent position in the feline TfR reduces but does not eliminate FPV binding [15] , [16] . However , in support of additional amino acid substitutions in the apical domain of TfR also being important , substitution of apical domain residue Leu221 in the feline TfR also reduce virus binding and cell infection by CPV and FPV [16] . Those studies raise questions about how TfR has evolved to modulate its propensity to mediate infection . However , little is know about the evolutionary history of the TFRC gene among animals in the order Carnivora , many of which are hosts to viruses closely related to the FPV . Two major suborders are present within the Carnivora , the Feliformia and the Caniformia , and those represent 16 families in total [17] . Confirmed or likely parvovirus infections have been reported for members of most families , although less commonly or not at all among pinnipeds [18] . Here we examined the diversity of TFRC gene sequences among some hosts distributed across the order Carnivora and find evidence for positive selection of this gene , more specifically in the Caniformia . Some variable sites , including some that were under positive selection , are located in the structural region of the apical domain that influences parvovirus binding . When some of these historical mutations were introduced into the apical domain of TfR they reduced binding by parvoviruses , making parvovirus a plausible selective force for the retention of these mutations when they occurred in nature . This suggests that there may have been viral pressure on this receptor before CPV emerged , some of which was exerted by ancient parvoviruses . A glycosylation site mutation present in the TfR of dogs that appears to protect dogs against FPV infection arose in a common ancestor of dogs and coyotes , suggesting that it is a recent change in that lineage .
Because of the recent emergence of CPV , it was previously thought that parvoviruses have been infecting Feliforms and some Caniforms for much longer than they have infected domestic dogs and closely related coyotes and wolves . We therefore wished to examine how the evolution of TfR reflected this history of infection in these different species . The sequences of the complete TFRC genes from 17 different carnivores , and of the apical domain from Otocyon megalotis ( bat-eared fox ) , were determined by cDNA sequencing or obtained from sequence databases ( Table 1 ) . These orthologous sequences differed by up to 10% at the nucleotide level , but were easily aligned so that patterns of non-synonymous and synonymous mutational accumulation could be analyzed . The dN/dS ratio captures the number of non-synonymous mutations present per non-synonymous site ( dN ) compared to the number of synonymous mutations present per synonymous site ( dS ) [19] . Most protein-encoding genes accumulate far fewer non-synonymous mutations than synonymous mutations ( dN/dS≪1 ) due to selective constraints [1] . In evolutionary arms race scenarios such as the ones that can develop between hosts and viruses , continued selection of each party for evolutionary dominance can cause accelerated evolution of proteins involved in the host-virus interaction . In such situations , the recurrent positive selection for non-synonymous mutations that provide a selectable advantage to each party results in relevant genes acquiring a dN/dS>1 signature [20] . Because the apical domain is the binding site for parvoviruses , we first analyzed the evolutionary history of this domain by calculating the dN/dS value on each branch of the tree ( Figure 1 ) . Surprisingly , given that CPV is thought to have been in dogs for less than 40 years , the branch leading to dog has the highest value of dN/dS on the entire tree ( dN/dS = 1 . 7 ) . On that branch the region of TFRC encoding the apical domain is estimated to have accumulated 8 non-synonymous and 2 synonymous changes since dog and red fox shared their last common ancestor 9 to 10 million years ago [17] , [21] , [22] . In contrast , the apical domain encoding region of the red fox TFRC has acquired 0 non-synonymous changes and 2 synonymous changes over the same time period . To investigate this further , we examined the apical domain region of the TFRC genes of other Caniform species closely related to domestic dogs , the coyote , black-backed jackal , and bat-eared fox ( Figure 1B ) . This provided increased resolution to the timing of the acquisition of the 8 non-synonymous mutations that separate the TfR apical domains of dog and red fox , showing that they accumulated along the long lineage leading to dogs ( red branches labeled 1–3 in Figure 1B ) . The specific non-synonymous TfR mutations predicted to have occurred along key branches in the genus Canis ( branches 1 , 2 , 3 ) are listed below the phylogenetic tree ( Figure 1B ) . These include K384N ( canine TfR numbering ) , which resulted in the novel glycosylation in the apical domain of the canine TfR which controls FPV binding [15] . Thus , the greatest rate of protein evolution observed in the apical domain occurred on the branches leading to dogs and their closest ancestors , species that are thought to have harbored parvoviruses for only around 36 years . The accelerated accumulation of non-synonymous mutations on the lineage leading to dogs could simply reflect relaxed selection on the apical domain of TfR in those species . In order to test the hypothesis that TFRC has evolved under positive selection for non-synonymous mutations , the sequences were fit to models that both allow and disallow some codons to have evolved under positive selection ( dN/dS>1 ) using PAML [23] , [24] . Null models were rejected in favor of models of positive selection in an analysis of all full-length carnivore sequences , both with and without the horse TFRC as an outgroup sequence ( p<0 . 001 for both analyses; Table 2 ) . In these analyses , between 9 and 14% of codon sites were assigned to a dN/dS class of approximately 2 . 0 , indicating that non-synonymous mutations have been fixed at a rate approximately twice that of synonymous mutations at these codon sites . These rapidly evolving codon positions are listed in the final column of Table 2 . We then analyzed the evolution of TFRC sequences from Feliform and Caniform species separately . In an analysis of the Caniform TFRC sequences ( 7 species ) , the null model could be rejected in favor of a model of positive selection ( p<0 . 001; Table 2 ) . Interestingly , although fewer codon sites were identified as evolving under positive selection ( only 5% of sites ) , those had a higher dN/dS value ( 4 . 0 ) than when Feliforms had previously been included in the analysis . When Feliform TFRC sequences were analyzed separately , the null model of could not be rejected in favor of a model of positive selection ( p = 0 . 23; Table 2 ) . Even though more Feliform sequences were analyzed than in the Caniforms-only analysis ( 10 versus 7 species ) , the Feliform species analyzed are less diverged from each other and thus the tree length of this dataset was only 0 . 55 ( Table 2 ) . The optimal tree length for PAML analysis has been shown to be ∼1 [25] , so the lack of support for positive selection in this group must be considered with that in mind . To formally test the hypothesis of Caniform-specific positive selection , we analyzed our full dataset with a “branch-site” model of evolution [26] to determine if there are codon positions evolving under positive selection specifically in the Caniform clade . This analysis supported caniform-specific positive selection ( p<0 . 006 , Table S1 ) , consistent with the higher dN/dS value for the class of positively selected codons that was observed when Feliforms were removed in the previous analysis . These data show that TFRC has evolved under positive selection during the speciation of Caniforms , particularly in species closely related to modern dogs . We next wished to test whether the variable sites in TfR affected parvovirus binding , and therefore whether ancestors of these viruses could have been responsible for driving a least some of the rapid evolution observed in TfR . Many positively selected codons were identified in the TFRC sequences ( Table 2 ) , and they were mapped onto the crystal structure of the human TfR [27] , [28] , [29] ( Figure 2 ) . We found seven residues under positive selection in the apical domain , which is the primary binding site for FPV and CPV ( Figure 2A ) . Interestingly , of those seven residues , three ( 379S , 218D , 304R ( canine numbering ) ) have experienced a non-synonymous mutation on the lineage leading to dog during the 9 to 10 million years since the last common ancestor of fox and dogs ( Figure 1B ) . Sites of positive selection are also identified on other parts of the protein , which may be the result of selection by other pathogens , extant or extinct . Indeed , in rodents , two distinct viruses , New World arenaviruses and the retrovirus MMTV , bind the TfR receptor on distinct interaction interfaces [28] , [30] , [31] . Outside the apical domain , three features were observed which contained residues under positive selection ( Figure S1 ) . First , several residues of the stalk region were under selection in Caniformia; O-linked glycosylation of this region regulates proteolytic cleavage of the stalk and release of a soluble ectodomain [32] , [33] . Second , the αI-3 helix , whose function is not yet known , contains a cluster of selected residues . Third , the αII-9 helix contains four residues under selection; mutation of this helix in a previous study reduced FPV infection but not binding [16] . The αII-9 helix lies under a disordered apical domain loop which has been implicated in parvovirus binding , and future structural studies could reveal whether these residues could be involved in binding to this virus . Signatures of positive selection were also detected in other , isolated residues . For example , a methionine at residue 635 in the helical domain is near residues involved in Tf and HFE binding , and canine residue 150 aligns next to human polymorphism S142G which is associated with type 2 diabetes [34] , [35] . Further studies would be required to assign a function to any residues under positive selection outside the apical domain . To address the possibility that an ancient arms race with parvoviruses has been responsible for the positive selection of TFRC , we tested whether the changes of the positively selected residues in the apical domain of TfR alter parvovirus binding . Three of the seven sites of positive selection are located in surface-exposed loop regions near the border of the apical and protease-like domains . One of these ( T300; feline coordinate ) was mutated in a previous study and reduced parvovirus binding and infection [16] . Four of the positively selected residues are located within two adjacent β-turns comprising the lateral tip of the apical domain , the βII-1 to βII-2 turn and the βII-7b to βII-8 turn ( Figure 2B ) . In previous studies , residues of the βII-1 to βII-2 turn were mutated to alanines with no effect [15] , [16] . Two sites under positive selection are located in or adjacent to the βII-7b to βII-8 turn . This turn is also close to the glycosylation site mutation ( residue 384 in the canine TfR ( 383 in the feline TfR ) ; black in Figure 2B ) , a critical determinant in controlling specific binding of FPV or CPV capsids . When this turn was mutated previously , many of the changes prevented cell-surface expression [15] , [16] . Because no information existed on the βII-7b to βII-8 turn , we focused our attention on this structural feature using evolutionarily-informed substitutions to solve the problem of expression . To test the functional affect of the observed mutations , mutations were made in the background of the feline TfR since this receptor can be utilized by both FPV and CPV [8] , [36] , so it can serve as a platform for testing Caniform-specific evolutionary adaptations . We mutated three residues within the βII-7b to βII-8 turn in the feline TfR , as predicted from homology modeling of the feline receptor onto the human TfR structure ( positions 378 , 379 , and 380 ( feline numbering ) ) to each three-residue combination found among the carnivore species for which we had sequences ( Figure 3 ) . Some of these recapitulate key mutations that were acquired by lineages in Feliformia ( Pallas cat , puma , and lion ) , while others were mutations acquired in lineages in Caniformia ( mink , fox , and coyote ) ( Figure 3 ) . After expressing each mutant TfR on TRVb cells , they were tested for binding of CPV and FPV ( measured at 4°C ) , or for binding and uptake ( measured at 37°C ) ( Figure 3A and B ) . In a previous study in CHO cells ( from which TRVb are derived ) , it was shown that holding cells at 4°C inhibited uptake in an assay modeling virus-cell fusion [37] . An antibody against the conserved cytoplasmic tail of the receptor was used to verify that each mutant TfR expressed to similar levels ( Figure S2 ) , and Tf binding was not significantly different between the various mutant TfRs ( results not shown ) . Interestingly , several of the mutational combinations tested showed reduced parvovirus binding and uptake , consistent with the idea that naturally occurring mutations could have been selected for this purpose ( Figure 3A and B ) . Mutations that represented combinations from Caniform species had bigger effects on binding to both viruses than those from other Feliform species . Importantly , this indicates that there are effects from mutational differences other than the glycosylation site ( Asn 384 in canine TfR ) that distinguish the Feliform and Caniform TfRs with regard to virus interactions . Caniform-specific mutations reduced binding , consistent with these mutations providing an adaptive advantage against virus infection . However , the binding patterns seen were similar for both CPV and FPV , indicating that these two viruses interact with a common region of the receptor , although this is perhaps not surprising given the these two viruses are >99% identical in sequence [38] . We also tested for FPV infection of cells expressing these mutant receptors , and only small differences were observed , from a minimum of 26% to a maximum of 34% ( Figure 3C ) , and the biological relevance of the differences is unclear . However , an ancient arms race between TfR and parvoviruses is also supported by the observation that the complementary region in the parvovirus capsid associated with receptor binding shows strong evidence of positive selection [38] , [39] . The N-linked glycosylation site in the apical domain of the canine TfR ( at position 384 in that sequence ) is critical for preventing FPV binding and infection of canine cells and dogs [36] . Among the carnivore species surveyed , only the domestic dog and coyote TFRC sequences encoded an Asn at this position , and we assume that it predates their speciation . However , the TFRC sequence of the closely related black-backed jackal does not possess this mutation . This allowed us to present the hypothesis that this single mutation could have been potent enough to end the arms race and prevent infection of the ancestors of dogs by parvoviruses for millions of years , until the emergence of CPV in 1978 ( Figure 1B ) . We therefore introduced the Lys to Asn change into codon 384 of the black-backed jackal TfR , which diverged just before the acquisition of this K384N and four other non-synonymous mutations found in dogs and coyotes . In a previous study in which TfR was expressed in TRVb cells , the Asn to Lys change was introduced into a wild type canine TfR background and resulted in a gel shift consistent with the loss of a glycan at this site , so the mutated jackal TfR should be glycosylated in this system [15] . As can be seen in Figure 1B , there are no mutational differences in the TfR apical domain between the black-backed jackal and the most recent common ancestor of this jackal and the domestic dog , so the jackyl sequence can be thought of as an ancestral representation of the apical domain as it existed before this glycan-introducing mutation appeared . Cells expressing wild-type jackal TfR bound both FPV and CPV capsids , and were also susceptible to infection by both viruses ( Figure 4 , Figure S3 ) . When the black-backed jackal TfR with the K384N change was tested , that showed significant reductions in FPV binding , uptake , and infection compared to the wildtype jackal or the feline TfR ( Figure 4 ) . However , these levels , while low , were still higher than seen for the wildtype canine TfR ( Figure 4 ) , indicating an additional role of other sequence changes in the canine TfR in controlling binding and infection by FPV . The K384N mutation in the jackal receptor did not affect CPV as much as FPV in any of the assays , and it is known that CPV successfully compensates for this novel glycosylation in the TfR to allow infection of dog cells [15] , [36] .
Here we show that there has been significant adaptive evolution of the host TFRC gene over the ∼54 million years of evolution of the members of the order Carnivora , and particularly in the Caniforms . One of the suggestions of this work is that parvoviruses with properties similar to those infecting hosts now were also infecting them millions of years ago , which has implications for parvovirus in the new field of “paleovirology , ” the study of ancient viruses [40] . Natural selection has sampled a number of mutations in TfR over evolutionary time , and by introducing these into the background of the feline TfR we showed that some of those likely modified parvovirus interactions when they occurred . Dogs , coyotes and wolves were not infected by the FPV-related viruses until CPV emerged in the mid-1970s , and we therefore expected that the TFRC orthologs from those hosts would show the lowest amount of variation in the region of the apical domain which contacts the parvovirus capsid . We were surprised to find that the lineage leading to dogs showed a relatively rapid acquisition of non-synonymous mutations , most likely during the period between about 9 and 3 million years ago [22] . One of these mutations introduced a novel glycosylation site at residue 384 in the canine TfR , and that receptor was subsequently able to resist binding and infection by the FPV-like viruses , as illustrated by the reconstruction of this evolutionary event in the background of the jackal receptor . The sequence of the black backed jackal was very close to that of the ancestor of the lineage leading to dogs , and therefore this was very similar to the event that arose in the common ancestor of these hosts . The appearance of the K384N variation in the lineage leading to dogs , wolves , and coyotes likely occurred around less than six million years ago , and that may have been a potent enough mutation to inhibit parvovirus infection and extinguish the arms race . CPV subsequently arose when an FPV-like virus acquired mutational changes that made it able to efficiently infect cells expressing TfR with the canid-specific glycosylation site [8] , [15] . Glycans play an important role in the biology of many pathogen receptors , and this study yields the new perspective that these glycans might sometimes be adaptively gained during host-pathogen arms races . One idea is that post-translational modifications such as glycans may provide physical distance between pathogen and receptor , explaining why mutations that introduce new glycosylation sites can be so potent . This also modifies our previous understanding of this well-known example of viral emergence , introducing the idea that canine parvovirus was a re-adaptation of the virus to the resistant receptor of a former host . Selection on the TfR likely resulted in both the reduced binding to FPV-like ancestors through the acquisition of non-synonymous mutations , and the complete resistance through acquisition of the novel glycosylation site mutation . That resistance was clearly overcome in the 1970s when the CPV-ancestor gained the small number of capsid mutations that re-established binding of the canine TfR , allowing the emergence of CPV as a new pandemic pathogen . Therefore , the CPV host-switching event was the re-adaptation of a pathogen that had previously infected the ancestors of dogs . One question is why CPV emerged only recently , given the length of time that dogs have apparently been resistant . The size of the dog population has increased significantly in the ∼10 , 000 to 20 , 000 years since they were domesticated , and it is possible that CPV-like viruses emerging before dog domestication would not have maintained sustained transmission up to the present . It is difficult to connect the host evolution occurring over geological timescales with the more rapid evolution of the viruses . This parvovirus model therefore provides a particularly clear description of both the host and viral sides of a long-standing interaction . An uncertainty in all studies of this type is a lack of knowledge of the viruses or other pathogens that were responsible for the selection that occurred millions of years ago . However , integrated viral sequences in various vertebrate genomes show definitively that related ancestral parvoviruses were infecting mammals millions of years ago [4] , [5] , [6] . While these viral sequences are 40–60% or more diverged from modern viruses in amino acid sequence , it is plausible that ancient parvoviruses could have bound the TfR and imposed the selection seen . Could an FPV-like ancestor have imposed a sufficiently strong pressure on a host population to select for variants of this key factor involved in susceptibility to infection and disease ? This appears to be likely based on the few studies that have examined the effects of FPV on wild populations , where losses of up to 90% of the young each year may be due to these infections [18] , [41] , [42] . After mutations reducing pathogen infection become widely distributed in a host population , the development of viral adaptation to those changes is to be expected , resulting in the development of an evolutionary arms race . For viruses this has been seen in the cases of retroviruses and the cellular factors that control their infection , and in the selection of viral-controlling immune properties of the hosts and their viral countermeasures . Although the rates of mutation of hosts and their viruses are many thousand-fold different , the complexity of the processes required to overcome the host changes may cause significant delays in the acquisition of the necessary combinations of mutations [1] . In the case examined here , it may have taken millions of years for the FPV-like viruses to overcome the virus-blocking adaptations in the canine TfR , indicating the complexities of the biological and evolutionary mechanisms involved in host shifting even when only a small number of changes in the virus are required .
We examined a total of 19 TfR sequences ( Table 1 ) . Of those , we determined the sequences of the TFRC gene open reading frame of 14 host species , as listed in Table 1 . In 13 cases , RNA was isolated from frozen primary cultures of cells , from frozen tissues , or from immortalized mink cells ( the CCL64 cell line ) using an RNeasy kit ( QIAGEN Inc . , Valencia , CA ) . RNA was isolated from Vulpes vulpes ( red fox ) tissue which had been frozen in TRIzol by the manufacturer's protocol ( Invitrogen , Carlsbad , CA ) . One-step RT-PCR was performed using SuperScript reverse transcriptase and Platinum Taq DNA polymerase ( Invitrogen ) , using primers in the 5′ and 3′ non-translated regions of the TfR mRNA . PCR products were purified using a QIAquick kit ( QIAGEN ) and either directly sequenced or cloned into the plasmid pCR-2 . 1-TOPO , and the cloned fragment sequenced . In addition to the 14 sequences determined here , we analyzed previously published Felis catus and Canis lupus familiaris TFRC sequences [8] , [43] as well as the Giant Panda ( Ailuropoda melanoleuca ) TFRC derived from the panda genome project [44] . The raccoon ( Procyon lotor ) TFRC sequence was obtained from a raccoon cell line and from a raccoon tissue sample [45] , while the horse ( Equus caballus ) TFRC sequence used as an outgroup was previously published [46] . The multiple sequence alignment generated for TFRC was analyzed for positive selection with the “codeml” program in PAML [24] . This offers several models for gene evolution , some where no codons are allowed to evolve with dN/dS>1 ( NSsites models M1a , M7 and M8a ) , and others where positive selection of some codons is allowed ( NSsites models M2a and M8 ) . A likelihood ratio test allows comparison of positive selection models to null models . TRVb cells ( Chinese hamster ovary cells , which do not express endogenous TfR ) were cultured in Ham's F12 medium with 5% fetal calf serum [47] . The black-backed jackal TfR was amplified as described above , cloned in the pCDNA3 . 1 ( − ) vector for expression . The feline and canine TfR constructs used are previously described [43] . Mutations were introduced into the jackal and feline TfRs using the Phusion mutagenesis protocol , as previously described [16] . Receptor expressing plasmids were transfected into TRVb cells seeded in 9 cm2 trays at a density of 2×104 cells per cm2 and transfected with 1 . 5 µg ( for infection assays ) or 2 µg ( for binding assays ) of TfR plasmid or pCDNA 3 . 1 ( − ) using Lipofectamine ( Invitrogen ) . Two days after transfection , TRVb cells were detached with trypsin/versene and seeded at a 1∶5 dilution on coverslips . The next day the cells were washed and incubated for one hour with FPV , CPV , or virus-free medium . Five days after transfection , cells were fixed in 4% paraformaldehyde and stained with mouse anti-human TfR cytoplasmic tail antibody ( clone H68 . 4 , Invitrogen ) followed by Alexa 488-conjugated goat anti-mouse secondary antibody to detect TfR , and then with Alexa 594-conjugated mouse anti-NS1 antibody [48] . Two days after transfection , TRVb cells were washed with cold Dulbecco's PBS and detached using Accutase ( Innovative Cell Technologies , San Diego , CA ) . Cells were pelleted and washed in PBS containing 1% ovalbumin . Cells were then incubated for one hr at 37°C or at 4°C with iron-loaded canine Tf conjugated to PerCP dye , and Alexa488-conjugated genome-free CPV-2 or FPV capsid . After washing with PBS with 1% ovalbumin , 10 , 000 cells were analyzed by a Guava EasyCyte Plus ( Millipore , Billerica , MA ) . Cells were gated based on forward and side scatter and compensated in FlowJo 9 ( TreeStar Inc , Ashland , OR ) . Tf-positive cells were gated in FlowJo and exported to JMP for statistical analysis because Tf labels the transfected cells expressing the receptors . All receptors are expressed to similar levels on TRVb cells , as revealed by staining with an antibody against the cytoplasmic tail of the receptor ( Figure S2 ) . Since Tf does not compete with parvovirus for TfR binding [15] , [16] and the mutations introduced are far removed from the TfR binding site , the relative Tf-PerCP fluorescence of cells reflected the expression of the receptor on that cell . For each cell , the fluorescence intensity of parvovirus binding was normalized to the fluorescence intensity of Tf . One way analysis of variance of the mean ratio of fluorescence intensities of each receptor was performed to determine the degree of parvovirus binding , or binding and uptake . Since the choice of the domestic cat TfR for this study is arbitrarily chosen as the background for the mutation analysis , Tukey's HSD was used to detect significantly different levels of binding and uptake instead of pairwise comparisons to the feline TfR . | Parvoviruses in cats and dogs have been studied as a model system to understand how viruses gain the ability to infect new host species . By studying the evolution of the transferrin receptor , which the virus uses to enter a cell , we discovered that the ancestors of dogs were likely infected by a parvovirus millions of years ago until they evolved and became resistant; this was caused by their transferrin receptor changing so it no longer bound the virus . When a variant virus that infects dogs emerged in the 1970s , it had adapted to overcome this block . This story suggests that diseases which were once eliminated from a species can evolve and regain the infectivity for that host , therefore having high potential to be emerging diseases . We identified features of the receptor that were important to the evolution of this host-virus interaction and confirmed their role in regulating virus binding in cell culture . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods",
"and",
"Materials"
] | [
"viral",
"attachment",
"viral",
"evolution",
"host",
"cells",
"viral",
"transmission",
"and",
"infection",
"virology",
"emerging",
"viral",
"diseases",
"biology",
"microbiology"
] | 2012 | Evolutionary Reconstructions of the Transferrin Receptor of Caniforms Supports Canine Parvovirus Being a Re-emerged and Not a Novel Pathogen in Dogs |
We have compared 12 genome-scale models of the Saccharomyces cerevisiae metabolic network published since 2003 to evaluate progress in reconstruction of the yeast metabolic network . We compared the genomic coverage , overlap of annotated metabolites , predictive ability for single gene essentiality with a selection of model parameters , and biomass production predictions in simulated nutrient-limited conditions . We have also compared pairwise gene knockout essentiality predictions for 10 of these models . We found that varying approaches to model scope and annotation reflected the involvement of multiple research groups in model development; that single-gene essentiality predictions were affected by simulated medium , objective function , and the reference list of essential genes; and that predictive ability for single-gene essentiality did not correlate well with predictive ability for our reference list of synthetic lethal gene interactions ( R = 0 . 159 ) . We conclude that the reconstruction of the yeast metabolic network is indeed gradually improving through the iterative process of model development , and there remains great opportunity for advancing our understanding of biology through continued efforts to reconstruct the full biochemical reaction network that constitutes yeast metabolism . Additionally , we suggest that there is opportunity for refining the process of deriving a metabolic model from a metabolic network reconstruction to facilitate mechanistic investigation and discovery . This comparative study lays the groundwork for developing improved tools and formalized methods to quantitatively assess metabolic network reconstructions independently of any particular model application , which will facilitate ongoing efforts to advance our understanding of the relationship between genotype and cellular phenotype .
Efforts to map metabolic networks—to describe the full network of anabolic and catabolic biochemical reactions occurring within a cell—have advanced from early biochemical studies of fermentation [1] to contemporary efforts to algorithmically generate pathway diagrams from genomic sequence [2] . Such pathway maps may be augmented with additional metadata to build a digital “reconstruction” of an organism’s metabolic network . In turn , such organism-specific reconstructed metabolic networks may be further supplemented to build mathematical models that are capable of simulating metabolic fluxes [3] . Recently , research efforts have focused on improving the ability to quickly build genome-scale metabolic network models of metabolism and to improve their predictive accuracy [2 , 4–6] . Comparatively less effort has been spent exploring opportunities for knowledge discovery arising during the process of network reconstruction prior to mathematical simulation [7] . In this work , we emphasize the distinction between metabolic network “reconstruction” and metabolic network “model” . Emphasizing this distinction facilitates an effort to resolve the relative contributions to model predictive accuracy or error arising from the metabolic network structure itself ( the “reconstruction” ) from those arising from mathematical parameters chosen when building a simulatable metabolic network “model” from the network reconstruction . While a variety of ad hoc quantitative metrics have been applied to evaluate improvements in metabolic network models , quantified assessment of the progress of the underlying reconstructions over time is a nascent effort [8] . This may be , in part , due to the fact that the number of models is so much greater than the number of extensively curated reconstructions . The relative difficulty of curating a comprehensive metabolic network reconstruction compared to generating a draft model is highlighted by the fact that there are currently more than 2 , 600 functional draft models [6] , but only Escherichia coli [9] and Saccharomyces cerevisiae [10] reconstructions have been extensively updated multiple times and revised through curation efforts by multiple research groups over a multi-decade timescale . Since simulation results are more amenable to quantitative analysis than reconstruction quality , reconstructions have generally been assessed indirectly , often in the context of model performance via manuscript discussion of scope ( the number of genes , reactions , or metabolites in a model ) , standards compliance , naming or annotation conventions , reputation of the group that built a particular model , or predictive performance of a derived model for a particular phenotype of interest ( commonly used phenotypes include gene essentiality , substrate utilization , growth rate , or product production ) [11] . This indirect approach to assessing metabolic network reconstruction quality bears a risk , however , because the model building process itself can obscure important details about the underlying reconstruction ( particularly knowledge limitations that may be useful for informing future investigation [12 , 13] ) . The standard reconstruction protocol includes converting a reconstruction to a mathematical model for subsequent debugging [3] . Thus , the ability to create a functional model has come to serve as a minimum threshold for defining the scope of a “draft reconstruction” , and the distinction between “reconstruction” and mathematical “model” has become blurred . Model developers are free to take different approaches when parameterizing model features such as objective function ( i . e . , biomass composition ) [14 , 15] , media definition [16] , and reference lists of “essential” genes used for benchmarking model performance [17] . Model developers may use different approaches to gap-filling [4 , 18 , 19] , trimming dead-end metabolites , establishing an objective function , and adding transport and exchange reactions . In fact , optimization-based approaches have been applied to successfully improve model essentiality predictions by adjusting these parameters [20] . Such algorithmic approaches can improve the predictive performance of a model even in the absence of any changes to the underlying metabolic reconstruction . Using model performance to drive iterative improvements to metabolic network reconstruction has led to two perverse consequences . First , if two models of the same organism give different predictions , how can a researcher determine whether the differences arise from differences in the reconstructed network or from differences in model parameters ? We have previously observed that algorithms such as OptKnock [21] can suggest different targets for metabolic engineering efforts when applied to different models of the same organism ( unpublished data ) . Second , a single metabolic network model can provide only limited information about the quality of the underlying metabolic network reconstruction because there are so many degrees of freedom associated with deriving a model from a reconstruction [7 , 8] . Comparative analysis of multiple models , now possible at scales not previously feasible [22] , provides an opportunity to address these challenges of single-model analysis . Our approach is to conduct comparative analysis of yeast metabolic network models that have been published in the past two decades , while controlling for differing modeling assumptions with a standardized model biomass function , media definition , and common sets of genes considered in the evaluations . An additional benefit of comparative analysis of models spanning a multi-decade timescale is the opportunity for evaluating model predictive performance on data that was not available at the time of reconstruction , which can provide a useful independent validation procedure and provide insights into the degree of overfitting possible in these models through the manual reconstruction process that is very difficult to ascertain otherwise . At least 25 models of the Saccharomyces cerevisiae metabolic network have been published since 2003 [5 , 11 , 17 , 20 , 23–40] . Each of these models has been applied successfully to research efforts focusing on advancing biotechnology [41] , mapping genotype to phenotype relationships in cellular physiology [42] , or developing new methods in computational biology [43] . Previously , researchers have combined comparative analysis of three of these models ( iFF708 , iLL672 , and iND750 ) with experimental data to refine characterization of cellular phenotypes in 16 environmental conditions [44] , and developed tools to facilitate model matching and comparison for synchronous investigation or building composite models [45] . Another two models ( Yeast 5 and iMM904 ) have been evaluated for predicting growth rates of a prototrophic gene deletion library in 20 different conditions [46] . More recent efforts have begun comparative analysis of a broader range of these models [22]; however , we are not aware of previous large-scale comparative analysis efforts that modify model objective functions , reference phenotype lists , and simulated media composition to evaluate the underlying metabolic network reconstructions built for S . cerevisiae . In this study , we conducted 161 in silico screens of predicted single gene essentiality using 18 different simulated media conditions , 12 different yeast metabolic models , and 13 different biomass definitions . We used this range of simulation parameters to standardize choices made in the development of various models , thus facilitating evaluation of the underlying network reconstructions . Using a binary growth/no growth assessment metric , we evaluated model predictions of the essentiality of three different lists of “essential” genes compiled from literature and database review . Additionally , we conducted simulations of aerobic growth with constraints on glucose , oxygen , and nitrogen exchange reactions singly or in combination to evaluate the correlation between model predictions of maximally achievable biomass flux values and reported experimental growth rates . We also conducted 10 in silico screens of pairwise gene essentiality by different models using their default media and biomass definitions , and compared model predictions to 32 , 488 gene pairs annotated as synthetic lethal in the Saccharomyces Genome Database . All code for our analysis is available as S1 File . Our key findings include the following . ( 1 ) Changes in model scope reflect a history of iterative reconstruction development via collaboration between groups—in other words , each model contains evidence of its history , with stylistic and content evidence of the specific model from which it is derived . Knowledge is propagated between models , but there is also risk of error propagation . Therefore , it is important to revisit assumptions made when earlier models were originally built when evaluating newer models . ( 2 ) Model updates tended to fall into two major categories , model scope expansion ( i . e . the inclusion of new metabolic processes ) or subsequent refinement ( i . e . including essentially the same sets of processes but working to improve accuracy ) . There was a pattern in analyzing the models’ ability to predict KO essentiality that accuracy on average reduced when model scope expansion was done and then improved on subsequent reconstructions aimed at improving the same set of processes . ( 3 ) For each model , single-gene essentiality predictions were affected by parameters external to metabolic network structure , such as simulated medium , objective function , and the reference list of essential genes . ( 4 ) The correlation between model predictions of maximum biomass flux correlate and reported growth rates are the same for all models when only a single exchange reaction is constrained , but the correlation between model prediction and reported growth rate differ among the models when multiple exchange reactions are simultaneously constrained to experimental values . This difference can be attributed to changes in metabolic network reconstructions independent of model parameters . ( 5 ) The predictive ability for single-gene essentiality did not correlate with predictive ability for our reference list of pairwise synthetic lethal genes . Thus , we conclude that the reconstruction of the yeast metabolic network is generally improving , and have demonstrated that comparative model analysis contributes to reconstruction improvement . There remains great opportunity for advancing our understanding of metabolic function through continued efforts to improve the reconstruction of the yeast metabolic network .
We compiled summary statistics for functional yeast metabolic models published since 2003 ( Fig 1 ) , including the number of metabolites , reactions , dead-ends , gene-associated reactions , and genomic coverage . When the models are ordered chronologically , none of these statistics demonstrates continuous improvement , perhaps reflecting the differing research objectives that motivated the development of each new model , but also demonstrating the limited ability of any individual statistic for fully describing model quality ( e . g . reducing the number of genes in a reconstruction is an improvement if previous iterations included misannotated genes , but such a reduction of genomic coverage could be considered a worse statistic ) . We observed a general increase in the number of genes included in models over time , but this increase was not uniform . We also found that increased genomic coverage could be a result of modelers including genomic features that are no longer considered genes . For example , the Biomodels . db model accounts for the greatest coverage of the yeast genome , but includes 28 open reading frames currently annotated as “dubious” , or unlikely to encode a functional protein . Additionally , increases in genomic coverage did not imply improved predictive accuracy: Yeast 6 includes fewer genes than Yeast 5 , but improves single-gene essentiality predictions Similarly , the number of metabolites and reactions and the proportion of dead-end metabolites in yeast metabolic models has generally , but not uniformly , increased over time—and does not coincide with improved predictive ability . For example , the number of metabolites and reactions in Yeast 7 is much larger than that in Yeast 6 , though they have the same overall MCC . In contrast , iND750 contains fewer metabolites and reactions than its progenitor model iFF708 . The portion of dead-end metabolites—those metabolites that are consumed but not produced in the network or vice versa—has also varied among models , and does not correlate with predictive accuracy . Next , we evaluated model scope by comparing genomic coverage and the metabolites that could be cross-identified with Chemical Entities of Biological Interest ( ChEBI ) identifiers with the annotation included with the models . We were unable to directly compare model reactions because of the lack of standardized reaction identification between models , and the lack of an external reaction reference database identifier in any yeast metabolic model , a current limitation for interoperability and comparison in our field . We found that models clustered in groups that reflect their historical development [10] , but these clusters differ between gene and metabolite comparisons ( Fig 2 ) . Models clustered in 4 groups when comparing genomic coverage: 1 ) Versions 4–7 of the Consensus Reconstruction; 2 ) iMM904 , iMM904bs , iAZ900 , and iTO977; 3 ) a looser cluster of iFF708 , iIN800 , and iND750; and 4 ) the Biomodels . db model . A row-aligned comparative table of genes in each model is included as S1 Table . Model similarity clusters differed when based upon ChEBI identifier-annotated metabolites , and the clusters were more tightly linked to the research group most closely related to the development of a group of models . When clustered by annotated metabolites , the resulting 5 groupings consisted of 1 ) Versions 5–7 of the Consensus Reconstruction; 2 ) iND750 , iMM904 , iMM904bs , and iAZ900; 3 ) iFF708 , iIN800 , and iTO977; 4 ) Version 4 of the Consensus Reconstruction and 5 ) the Biomodels . db model . Improving model ability to predict the essentiality of individual genes for growth has not been the primary motivating factor for developing each new yeast metabolic network model , but essentiality predictions have generally been reported with the publication of each new model to demonstrate their accuracy and utility . However , direct comparison between reported predictive values is complicated by differing simulation conditions . In this study , we did not find a strong trend of continuing improvement in model ability to predict single gene essentiality over time . Instead , we found evidence of an iterative process in which model scope changes ( typically leading to a decrease in average predictive accuracy ) , followed by subsequent curation leading to improved prediction by descendant models ( Fig 3 ) . We found that both iIN800 , with its expansion of the reconstruction of lipid metabolism , and iND750 , with its expansion of compartmentalization , had a lower overall Matthews Correlation Coefficient ( MCC ) for single-gene essentiality predictions than their progenitor model , iFF708 . Subsequently , iMM904 refined iND750 , and made more correct predictions of single gene essentiality . Similarly , Yeast 6 refines Yeast 5 and improves predictive ability , but the focus of Yeast 7 on expanded scope does not lead to as large an improvement in single-gene essentiality predictive ability , and iTO977’s focus on expanding model scope to cover some protein modification processes and to provide a scaffold for integrating transcriptomic data does not lead to an improvement in predicting single-gene essentiality compared to its progenitor model , iIN800 . The Biomodels . db model was generated in a methods-development effort to improve automated reconstruction and annotation . The algorithm underlying the Biomodels . db model prioritizes connectivity and defines “functionality” as the ability of the model to predict growth using a generic biomass definition objective function . The Biomodels . db annotates genes with a different nomenclature than the ORF format used by other models , so was incompatible with our gene essentiality screen . We evaluated the Biomodels . db model by other comparative metrics , but did not evaluate its FBA performance here . We conducted 161 simulated genome-wide single gene deletion screens for gene essentiality by conducting flux balance analysis with an objective of maximizing biomass flux . We used different combinations of simulated media and biomass objective functions and compared model predictions to appropriate reference lists of essential genes , as described in the Materials and Methods section . We found that no single model predicted essential genes best in all simulations ( Fig 4 ) . In our simulations , the iAZ900 model had the highest single MCC we calculated ( 0 . 83 ) for a particular condition , and the Yeast 7 model had the highest overall MCC across all the conditions for which we calculated ( 0 . 61 ) . As a point of comparison , we calculated a MCC of 0 . 61 based on the reported results of a gene essentiality screen with a recent model of the E . coli metabolic network [47] , which is , along with yeast , widely considered the best studied genome-scale metabolic network model to date . Although it had the highest observed MCC in one condition , the iAZ900 model did not perform as well in other simulations—it also had the lowest MCC ( 0 . 17 ) for an out-of-sample screen using the iFF708 biomass definition , a very permissive set of exchange reactions , and a reference gene list based upon SGD-reported phenotypes . When the exchange reactions are constrained to reflect a simulated glucose minimal defined media , the iAZ900 MCC for the iFF708 biomass increases to 0 . 55 . Such ranges of model predictive ability were observed for all models across differing simulation conditions , highlighting the importance of controlling for model parameter variation when attempting to compare metabolic network models of a particular organism . In the specific case of iAZ900 , the excellent performance of its best condition reflects the authors’ goal in developing iAZ900 –to use an algorithmic approach to improve the iMM904 model by maximizing agreement with a list of genes essential genes reported to be essential by Kuepfer et al . [17] . The reference list of essential genes used in the development of iAZ900 originates from a screen of non-essential genes in the yeast knockout collection in glucose-limited defined medium [48] . This reference list for training the algorithm is one of the reference lists we used for comparative evaluation . iAZ900 did not perform as well at classifying genes as essential when using other reference gene lists . Thus , iAZ900 demonstrates that high model performance can be achieved by one metric , but there is the usual tradeoff between sensitivity and specificity when attempting to generalize a specific metabolic network model to predict phenotypes in new conditions . Our observation that model performance was influenced by the reference list of genes considered essential when attempting to evaluate model predictive ability demonstrates that the definition for gene essentiality is another parameter that may be tuned as model developers refine their model . In our simulations , model MCC was higher on average when calculated relative to the SGD-based list of essential genes for five of the models ( iND750 , iIN800 , iTO977 , Yeast 6 , and Yeast 7 ) , and higher relative to the Kuepfer-based list of essential genes for the remaining six models ( iFF708 , iMM904 , Yeast 4 , iAZ900 , iMM904bs , and Yeast 5 ) . These two groups do not correspond to the clusters identified when comparing model genomic coverage or the clusters identified when comparing annotated metabolites . All models predicted gene essentiality better when glucose was the simulated primary carbon source than when galactose , glycerol , or ethanol were the primary sources . However , since the reference gene list used for the non-glucose carbon sources was based upon a single screen , we could not determine whether this reflects limitations in the reconstruction of the non-glucose metabolic network , or strain and laboratory-specific effects in the reference data . Historically , the metabolism of non-glucose carbon sources has received less biochemical characterization than glucose metabolism in yeast . Since the objective function is a tunable parameter that is independent of metabolic network structure , we normalized the objective by selecting a biomass definition that each model could satisfy , as described in the Flux Balance Analysis—Biomass Definition subsection of Methods , below . Thus , we began differentiating between model parameter improvements and network structure improvements to compare the reconstruction underlying different models more directly . We performed Flux Balance Analysis of the metabolic network models using both the biomass definition provided by the model authors , and the biomass function used for the iFF708 model , and found that for all models with different biomass definitions than the iFF708 model , the model predictive power was affected by the objective function used ( Fig 5 ) . In every case but the Yeast 4 model , model predictions were better using the model default biomass objective than the iFF708 objective , suggesting that model developers have achieved improved predictive accuracy in part by modifying the objective function , and such improvements have been achieved independently of refinements to the biochemical network reconstruction itself . This approach is not meant to imply that modifications to an objective function would be conducted solely to improve a predictive metric: refinements to the biomass definition also reflect improved measurement of biomass composition and changes to model scope . We selected a common biomass definition for our analysis to evaluate the impact of this particular model parameter . We conducted FBA-based comparison of media- and objective-normalized model predictions of maximum achievable biomass fluxes with the aerobic growth rates reported by Österlund et al . for “N-limited” and “C-limited” conditions ( we did not simulate anaerobic growth since most of the models we are examining do not predict anaerobic growth on a minimal medium ) . We found that the model predictions of maximally achievable biomass flux correlated with the previously reported “N-limited” growth rates with a correlation of 0 . 994 when nitrate or nitrite exchange fluxes were constrained to previously reported uptake rates ( S3 Table ) . The “C-limited” simulations reflected a different behavior . When we constrained the glucose exchange reaction alone , all models had a 0 . 816 correlation with the reported growth rates ( Fig 6A ) . However , the growth rates labeled “C-limited growth aerobic” by Österlund et al . are not linear over the range of constraints imposed on the glucose exchange reaction , suggesting that carbon ( glucose ) flux is not the sole growth-limiting factor , particularly at the higher range of glucose flux constraints . The ratio of glucose exchange flux to oxygen exchange flux would be expected to strongly influence maximum achievable biomass flux due to stoichiometric constraints on the oxidation of glucose [49] . We tested model behavior against this expectation by conducting FBA with both glucose and oxygen exchange reactions constrained to values reported by Österlund et al . [37] . When glucose and oxygen exchange reactions were both constrained to experimental values , we observed that the models segregated to 2 groups: biomass flux predictions made by 7 models ( iFF708 , iIN800 , Yeast 5 , iTO977 , iMM904 , and iMM904bs ) correlated with observations with a correlation >0 . 9 , and predictions made by the remaining models ( Yeast 4 , Yeast 6 , Yeast 7 , iAZ900 ) had lower correlations ( Fig 6B ) . We used one-norm minimized FBA [50] to find an explanation for this difference in model predictions and observed unrealistically large fluxes through internal reactions along with unusually large exchange fluxes in models that overpredict biomass flux in high glucose:oxygen growth simulations . Through repeated FBA and manual investigation of high-flux loops , we found that the low-correlation models all had a flux through a mitochondrial aspartate transport reaction . This reaction is not associated with a gene in the iAZ900 model ( reaction id “ASSPt2M” ) , but is annotated with yeast open reading frame YPR021C in the Yeast 4 , Yeast 6 , and Yeast 7 models ( reaction ids “r_1163” , “r_1117” , and “r_1117” , respectively ) . YPR021C encodes Agc1p , a protein that “fulfills two functions… glutamate transport into mitochondria … and … aspartate-glutamate exchanger within the malate-aspartate NADH shuttle” [51] . Subsequently , we also found this reaction in the iND750 ( “ASPt2M” ) , iIN800 ( “AGC1_2” ) , iMM904 ( “ASPt2m” ) , iMM904bs ( “ASPt2m” ) , Yeast 5 ( “r_1117” ) , and iTO ( “AGC1_2” ) models . We did not find this reaction in iFF708 or the Biomodels . db models , which do not include mitochondria as a separate compartment . We did not find literature support for including yeast mitochondrial aspartatate transport as reconstructed in these models . Thus , investigating erroneous predictions of maximum biomass flux by four models at simulated high glucose:oxygen flux states allowed us to identify a reconstruction error common to all compartmentalized models , an error that is independent of model parameters . When we removed this reaction from the models , we found that it did not affect the predictions for the high correlation models , and improved all remaining correlations to >0 . 9 , with the exception of the Yeast 4 model , which still over-predicted the maximum biomass flux at high glucose:oxygen exchange constraint ratios ( Fig 6C ) . Using the models as distributed ( i . e . , with tuned biomass definitions and default exchange reaction constraints ) , we conducted a simulated screen of all pairwise deletions for 10 models ( iFF708 , iND750 , iIN800 , iMM904 , Yeast 4 , iAZ900 , iMM904bs , Yeast 5 , iTO977 , and Yeast 6 ) . Using a strict definition of synthetic lethality in which neither gene is individually essential , but are pairwise essential for growth , we found that the MCC for model prediction of synthetic lethal gene pairs ranged from 0 . 04 to 0 . 12 , when compared to a list of synthetic lethal gene pairs that we generated using the Saccharomyces Genome Database Yeastmine tool [52] ( Fig 1 ) . Additional summary statistics of these screens are included in Fig 7 . Surprisingly , we found that the MCC for synthetic lethal interactions did not correlate with the MCC for single-gene essentiality ( R2 = 0 . 0253 ) . The relatively low predictive ability of these metabolic network models for synthetic lethal gene pairs may be attributed in part to the fact that the reference list of synthetic lethal gene pairs is not well-established due to the challenge of conducting pairwise gene deletion screens in vivo [53] , the fact that predictions of multiple perturbations to a genetic network require more complex analysis [54] , and synthetic lethality phenotype observations may be greatly influenced by experimental design [16] . We anticipate that evaluating and improving constraint-based phenotypic predictions of multiple-gene deletions will advance hand in hand with efforts to experimentally explore gene interaction networks . The source code used for conducting these simulations is included as S1 File . Detailed results of all simulations conducted , including lists of true and false predictions for each model in each simulated single-knockout screen , are attached as S2 File .
When we directly compare functional models of the yeast metabolic network published over the past two decades , we note generally increasing trends in genomic coverage ( using either verified or total number of open reading frames annotating model reactions ) , number of metabolites , and reactions . We do not discern strong trends in number of dead-end metabolites in models , the percentages of reactions associated with genes , or in predictive accuracy for synthetic lethal genetic interactions . When comparing flux balance predictions with model default or standardized simulated media , objective functions , and reference lists of “essential” genes , we find that predictive power for single-gene essentiality has gradually improved . However , the observed trends are not uniform across all models and simulations we conducted . Each model has its own strengths and weaknesses; as demonstrated in a previous comparative study , “different models may be preferable for use in different applications” [44] . The uneven progress in improving model performance metrics reflects the historical path of iterative model refinement . Different models are developed to address different research questions , and are not necessarily focused on improving gene essentiality predictive accuracy . Thus , each new model may advance ( or regress ) when compared to previous models depending on the metrics used to assess the model . This is particularly evident when examining the relative performance of single gene essentiality predictions ( Fig 3 ) . For example , iND750 greatly expanded compartmentalization in the yeast metabolic reconstruction , but had lower predictive ability for single-gene essentiality than the earlier iFF708 model in our analysis . Development of the iAZ900 model demonstrated the utility of optimization-based procedures for improving model prediction , so it has the highest MCC for the reference conditions used for model development , but not the highest overall MCC across all conditions . The iIN800 model expanded the reconstruction of lipid metabolism , and the iTO977 model expanded the scope of yeast metabolic models to facilitate transcriptomic analysis . As new models integrate and improve upon earlier models , a path dependency on previous modeling or reconstruction efforts emerges . This necessary relationship can lead to iterative improvement , but can also propagate errors and complicate assessment of the reconstruction of the yeast metabolic network . Further , as models have been developed , different research groups have used different tools to validate their models ( such as different lists of genes reported to be essential in a particular strain background or experimental condition ) . Thus , no model should be considered “best” or definitive for all applications . Examining simulations across multiple models may be a prudent approach for building confidence in predictions . The results of our comparison of predicted maximum achievable biomass flux to measured growth rates emphasize that model users must take great care when imposing multiple constraints prior to conducting FBA , or when interpreting experimental growth rate measurements as being attributable to a single limiting nutrient . If a model user is attempting to compare simulated predictions with observed growth rates that scale linearly with the concentration of a single limiting nutrient , our results suggest that they should test to ensure that the model is operating within a linear range in which the desired nutrient is in fact the sole factor limiting predicted flux to the objective . If operating within such a regime , users could confidently scale FBA results by either varying model parameters ( such as ATP maintenance demands ) or by simple linear transformation of objective values found via FBA . However , model users must be wary of discontinuities arising from shifts in the limiting nutrient ( such as from glucose to oxygen ) . The high correlation between predicted biomass flux and observed growth rates in high glucose:oxygen exchange constraint ratio regimes is surprising not only because “normal yeast mitochondrial structures are disrupted when glucose levels are high” [55] , but because c . 2 , 000 genes are regulated by the diauxic shift [56] . These changes are dependent upon concentration , rather than flux [57] . Thus , it is likely that there are sets of constraints that should be applied to a metabolic network for condition-specific modeling . We did not observe that such constraints were necessary for predicting the stoichiometrically-determined shift from glucose-limited to oxygen-limited maximal growth rate over a range of glucose to oxygen exchange flux ratios . Differentiating universal constraints ( such as chemical stoichiometry ) from condition-specific constraints appears to have great potential as a fruitful avenue for future research efforts . We found that model gene essentiality predictions are biased by factors that are not reflective of the accuracy or completeness of the metabolic network reconstruction . Such factors include reference gene lists , choice of objective function for flux balance analysis , and simulated media used for in silico screens . However , it is likely that standardizing these factors ( as we have done in this study ) for comparing models is not sufficient for assessing the quality of the metabolic network reconstruction; model builders must make other choices when developing a model that is amenable to simulation from a network reconstruction . For example , since different models use different approaches to fill gaps in the known metabolic network or to ascribe catalytic function to a poorly characterized yeast genes , different models are likely to include different hypothetical transport or biochemical reactions with different levels of evidence or confidence in the accuracy of the functional role of a protein . Reconstructing a metabolic network provides an opportunity to highlight areas of uncertainty to productively guide future research efforts . This opportunity is distinct from the utility of mathematical simulation of fluxes using metabolic network models . In fact , deriving a metabolic network model from a reconstruction can obscure the knowledge gaps or uncertainty that can be highlighted during the process of network reconstruction [7] . This risk is particularly acute where poorly understood portions of metabolism are not clearly implicated in the research to which a model is being applied , or when highlighting knowledge gaps or ambiguity may hurt model performance according to metrics used to assess predictive performance or scope . For the yeast models we compared , some of which use current annotation and model exchange protocols and formats , there is no mechanism for a model user to identify knowledge limitations discovered during reconstruction of the underlying network , nor is there sufficient annotation describing the specific techniques used to address such limitations when the model was constructed within the published model itself . The current state-of-the-art for metabolic network modeling presents a significant barrier to entry for researchers who are not familiar with the idiosyncrasies of each model because these idiosyncrasies are not sufficiently documented within the model structure itself . Thus , though we observed that model predictions of gene essentiality are generally better for models evaluated with a simulated medium containing glucose as the primary carbon source than model predictions when using ethanol , glycerol , or galactose as a carbon source , we cannot conclusively attribute the improvement in glucose-essential prediction to improvements in the reconstruction of the biochemical reaction network because there is no clear mechanism for separating the information contained in the underlying reaction network reconstruction from the modeling assumptions and choices made in deriving a particular metabolic network model . Similarly , we cannot conclusively attribute the relative lack of improvement in predictions with non-glucose carbon sources among models to errors in the reconstruction rather than faulty model assumptions , idiosyncratic objective function definition ( i . e . , model overfitting ) , or biological factors such as condition-dependent gene essentiality for genes included in the reference list of “essential” genes . Selecting appropriate data sets for model validation presents an additional challenge to the reconstruction effort . Specifically defining the media and conditions in which a given gene ( or combination of genes ) is essential remains an ongoing and important area of research to advance our understanding of metabolism . In the absence of well-defined reference phenotypes , we cannot confidently ascribe the low predictive ability for pairwise essentiality to errors in metabolic network reconstruction , uncertainties in synthetic lethal phenotypes , or physiological processes which are not metabolic , such as gene regulation or cell cycle checkpoint events . Further evaluating and improving model predictive performance for conditional essentiality will be greatly assisted by use of new prototrophic yeast strains and genetic screens in specifically designed media [46] . Despite these methodological challenges , there is benefit to comparing metabolic network models for the same organism for filling gaps and for identifying mistakes and opportunities for further expansion of the metabolic network reconstruction . We note , for example , that iron metabolism is important to mitochondrial function , but is not included in these models . None of the models include folate , chitin , or hypusine in the biomass definition , a model building choice that leads to false negative gene essentiality predictions and dead-end metabolites , but also highlights opportunities for expanding the reconstruction of the yeast metabolic network . Similarly , since most models have been validated with laboratory results from strains originally designed to facilitate genetic investigation ( strains which bear auxotrophic markers in their genetic backgrounds ) [48] , it is likely that the reconstruction of portions of the yeast metabolic network ( such as nitrogen and sulphur metabolism ) is incomplete . Updating the reconstruction in support of research with a new prototrophic yeast mutant library [46] provides an exciting opportunity for refining our understanding of yeast metabolism . As different groups refine yeast metabolic network reconstructions and models , there should be a convergence to a full , accurate reconstruction of the complete network . We do not observe evidence that supports marked changes in the reconstruction , such of a marked shift in model predictive ability or genomic coverage in our analysis of models published to date . Further , recent work has observed that many enzymatic functions are not included in existing models and reconstructions [8] . Thus , the effort to reconstruct the yeast metabolic network is incomplete . Increased efforts to expand the scope of reconstruction , such as including signaling and regulatory network processes , may provide a way to advance efforts to reconstruct organism-specific networks . Our analysis suggests that metabolic network reconstruction efforts could benefit from emphasizing the distinction between reconstruction of known or hypothesized metabolic function , and metabolic models developed for particular applications . A reconstruction may be improved , but model performance may drop by some metrics ( for example , adding a parallel metabolic pathway could lead to false negative gene essentiality predictions in the absence of regulatory constraints blocking an available network branch , or conversely , adding condition-specific regulatory constraints could hinder predictive value for conditionally-essential genes in other environments ) . Similarly , model performance could be enhanced in some cases by removing established biochemistry ( and such a choice would be defensible if modeling a particular environment in which a portion of the metabolic network was unavailable due to regulation ) . Thus , we find that no single metric we used to compare metabolic network models is sufficient to evaluate the progress of the yeast reconstruction efforts . Models should be assessed by gene essentiality predictions , as well as the extent of evidence and annotation for included information , the size of the network , and network connectivity metrics . Unfortunately , current methods for annotating the workflow of model development makes such analysis challenging . In some cases , erroneous model predictions have been computationally corrected through changes that cannot be annotated in exchange formats . Thus , they become obscured , rather than highlighted in a way that would better facilitate further investigation . Similarly , although great efforts have been expended to assess the evidence for information in the published models , none of the SBML files we evaluated included confidence scores or full annotation of literature sources , so these assessments remain internal to a development group and are not effectively propagated to subsequent model users . This is in part a historical artifact—many existing standards such as SBML are intended to distribute models , rather than fully annotated reconstructions . Efforts such as the definition of the Pathway Tools schema [58] lay important ground work towards broader community participation in improving the process of metabolic network reconstruction and metabolic model derivation . Though reconstructing metabolic networks has been the focus of biochemistry for more than a century , computational metabolic network reconstruction is still a young field with great contributions to make . Through this comparative analysis of yeast metabolic network models , we hope to contribute to the ongoing efforts to improve our understanding of metabolism through collaborative network reconstruction , and to highlight opportunities for improving the process of metabolic network reconstruction and model derivation .
Models were obtained from public repositories , supplemental information , or research collaborators and modified as follows: Each of the models is provided in S1 File , formatted as . mat files containing the COBRA Toolbox data structure with any modifications to enable simulation . Different sets of genes have been observed to be essential for growth in different conditions , and different lists have been used for previous evaluations of model predictive accuracy . Therefore , we generated our own reference lists for the current comparative analysis . We began with the list of 1 , 120 unique open reading frames annotated as essential by the Yeast Deletion Project ( available at http://www-sequence . stanford . edu/group/yeast_deletion_project/downloads . html ) , then removed YCL004W and YKL192C , based upon literature review [64 , 65] . Since this list was generated from experiments using a complete medium , we used it as our reference for simulations of growth in a synthetic complete medium ( our approach to defining simulated medium is described in the “Flux Balance Analysis—Medium Definition” section below ) . For evaluating simulations of growth in a glucose-limited minimal medium , we supplemented the essential gene list with 441 additional open reading frames reported to induce auxotrophy upon deletion . This list was generated by downloading a list of ORFs annotated as auxotroph-inducing in the Saccharomyces Genome Database , removing those already included in the essential list and temperature-sensitive inositol auxotrophs , and modifying the list based on literature-based curation as described in the testYeastModel . m file , which is included in S1 File . Combining the essential ORF list with the auxotroph-inducing list resulted in a list of 1560 open reading frames as our reference list of genes considered essential in a minimal medium . We also compiled reference lists of essential genes based on the screen of non-essential genes in the knockout collection in defined media with different carbon sources conducted by Kuepfer et al . [17] . This screen evaluated 4869 open reading frames included in the yeast knockout collection . Applying the stringent standard of a score of 0 for ORF essentiality , we classified 59 ORFs as essential in defined medium with glucose as the sole carbon source , 307 with galactose , 291 with glycerol , and 332 with ethanol . When comparing model predictions for these medium , we did not evaluate all ORFs in each model , but instead only characterized the subset of genes in the model that were also evaluated in the Kuepfer et al . screen . These lists of essential genes are included in the testYeastModel_kuepfer . m script , which is included in S1 File . We built a reference list of synthetic lethal genetic interactions ( pairs of ORFs that are not individually essential , but become essential when both are deleted ) by querying the Saccharomyces Genome Database with YeastMine [52] . We built an XML query to search for interacting genes where the experiment type was listed as “Synthetic Lethality” . The resulting report was downloaded and imported to MATLAB to generate a list of 32 , 488 pairs of ORFs that have been annotated as synthetic lethal . The specific XML query is described in the analyze_double_results . m script , which is included in S1 File . We note that any reference list of gene essentiality is dependent upon experimental conditions , so different researchers may construct such lists in different ways . The predictive accuracy of any model is a function of the standard used , so different reference lists are expected to affect the specific MCC value of its agreement with observation . Thus , it is particularly important that the same list of “essential” genes or gene pairs be used when comparing different models . We conducted flux balance analysis of each model using both the model-default simulated medium composition , along with media formulations we defined in an effort to standardize model predictions . We defined the following media for our simulations: a minimal medium that enabled predicted biomass production for all the models in which glucose is the sole carbon source; a synthetic complete medium with glucose as the sole carbon source , which was based on previous computational screening efforts [66]; and the synthetic medium defined by Kuepfer et al . , using glucose , galactose , glycerol , or ethanol as the sole carbon source . The minimal medium was simulated by allowing unconstrained exchange of ammonia/um , oxygen , phosphate , sulphate , and setting a constrained uptake of glucose . The iMM904bs and Biomodels . db models did not predict growth using this medium when using their default biomass definitions ( biomass definitions are described below in the “Flux Balance Analysis—Biomass Definition” section ) . To enable FBA , the simulated minimal medium for these models was supplemented: the iMM904bs model required iron exchange , and the Biomodels . db model required the amino acids L-tyrosine , L-lysine , L-isoleucine , L-arginine , L-histidine , L-methionine , and L-tryptophan . The code used for setting the medium for each model is included in S1 File in the testYeastModel . m and testYeastModel_kupefer . m scripts . We conducted FBA for all models except the Biomodels . db model with two different objective functions: first , we used the model’s default biomass definition , as included in the published version of the model; second , we used a common biomass definition as similar to the iFF708 biomass definition as each model’s exchange reactions and metabolites allowed . The Biomodels . db model did not predict growth with the iFF708 biomass definition , so we only used its default objective function to verify functionality . We selected the iFF708 biomass definition as a reference standard because it was the objective function that most models could satisfy . For example , iFF708 would not be able to satisfy the Yeast 5 biomass definition due to the expanded sphingolipid requirement in the latter . Our use of the common , older biomass definition was intended in part to separate model improvements that arise from improved reconstruction from those that arise due to a more specific biomass definition . The code used to set the biomass definitions for each model is included in S1 File in the testYeastModel . m and testYeastModel_kupefer . m scripts . Accounting for the seven media compositions and two biomass definitions described above , we conducted flux balance analysis to predict single-gene deletion growth phenotypes for each model in fourteen different conditions . We elected to use used a tight threshold of binary growth/no growth prediction when comparing model growth predictions to our reference lists of essential genes because flux balance analysis of metabolic network models may be less predictive for mutant growth rates than for a binary essential/non-essential gene classification [46] and because growth rate predictions may be tuned by adjusting model parameters such as ATP maintenance reaction demands or constraints on carbon source utilization reactions . For this study , a gene was considered to be predicted as essential only if flux balance analysis of a simulated mutant predicted a maximum flux to the biomass objective of less than 1 x 10−6 flux units . The agreement between model gene essentiality predictions and the reference lists was quantified using the Matthews’ Correlation Coefficient ( MCC ) ( eq 1 ) [67] , a metric that considers true positive , true negative , false positive , and false negative predictions without any assumption of the frequency of observations in the reference dataset . MCC ranges from -1 ( when model predictions are the exact opposite of the reference dataset ) to +1 ( when model predictions match the reference data set ) . MCC= TP×TN−FP×FN ( TP+FP ) ( TP+FN ) ( TN+FP ) ( TN+FN ) ( 1 ) Where true positives ( TP ) , true negatives ( TN ) , false positives ( FP ) , and false negatives ( FN ) are defined as in [17]: a true positive prediction is one in which the model predicts that a gene is not essential for biomass production , and the gene has been annotated as not essential . Values for each confusion matrix , along with lists of positive and negative predictions , are included as S2 File . We also assessed model prediction of synthetic lethality , or double-gene deletion phenotypes , for 10 of the models . When comparing model predictions to the reference gene list , we defined true positive as predictions in which neither gene in a reported synthetic lethal pair is predicted to be essential by itself , and the pair is essential . We defined true negatives as predictions in which neither gene is predicted to be essential by itself , the pair is not predicted to be essential , and the pair is not reported to be synthetic lethal . We defined false positives as predictions in which neither gene is individually predicted to be essential for growth and the pair is predicted to be essential for growth , but the pair has not been reported to have a synthetic lethal interaction . We defined false negative as predictions in which neither gene is individually predicted to be essential and the pair is predicted to be non-essential , though a synthetic lethal interaction has been reported . We categorized incorrect predictions of single-ORF essentiality as “other errors”–such errors were not included in our MCC calculation , since they were accounted for in the in silico single gene knockout screen . We did not modify the models’ biomass definition or simulated medium composition for our double knockout simulation . We also note that our definition of synthetic lethal interactions , which requires a model prediction of greater than 10% of the predicted wild-type biomass flux , is an arbitrary , but strict requirement . It is likely that the MCC for synthetic lethal predictions would be influenced by the choice of minimum biomass flux , and we selected 10% as a representative example for this particular analysis . If slow-growing double mutants are scored as synthetic lethal in an in vivo screen , and included in our reference list of synthetic lethal pairs , a correct model prediction of low biomass flux could be scored as false negative . Blocked reactions are reactions that cannot carry a flux in a given simulation condition; thus , the number of blocked reactions may change for a given model with different biomass definitions or different allowed exchange reactions . We used the fastFVA module [68] to count the number of blocked reactions for each model when all exchange reactions were allowed to carry flux , and using both the model default and the iFF708 biomass definitions . Dead-end metabolites are metabolites that either participate in only one reaction , or can only be produced or consumed . Thus , they are a network feature that is not influenced by exchange reaction or biomass definition changes . We counted the number of dead-end metabolites in each model with the COBRA Toolbox detectDeadEnds function . The code used for blocked reaction and dead-end metabolite analysis is provided in S1 File . Model-specific lists of blocked reactions and dead end metabolites are included as S2 Table . Similarity of genomic coverage among models was assessed by hierarchical clustering based on pairwise distance of binary vectors of logical values for open reading frames included in a model’s reaction annotation ( i . e . , 1 if a given ORF is included in a model , or 0 otherwise ) . The binary vectors are presented as a heat map , and clusters are presented as a clustergram and scatterplot ( generated with classical multidementional scaling ) in Fig 2 . Different model developers have annotated metabolites in different ways , so we began our comparison of metabolites by expanding the annotation of models by adding identifiers from the Chemical Entities of Biological Interest ( ChEBI ) database [69] to metabolites where possible . We were able to establish ChEBI annotation for different subsets of metabolites in each model , so this comparison is , by necessity , less comprehensive than comparison of model genomic coverage . The Biomodels . db model annotates metabolites with multiple ChEBI identifiers ( reflecting redundancy in the ChEBI database ) . We chose the first ChEBI identifier when comparing the Biomodels . db model with models derived from manual reconstruction . Other models did not include multiple ChEBI identifiers for annotated metabolites . Like genomic coverage , metabolic coverage was scored with a binary vector of logical values , and the comparison is presented as a heatmap , clustergram , and scatterplot . A sorted list of genes by models and all code used for scope comparison are included as Supporting Information . We compared predictions made by media- and objective-normalized models with the aerobic growth rates reported by Österlund et . al . [37] for “N-limited” and “C-limited” conditions ( we did not simulate anaerobic growth since most of the models we are examining do not predict anaerobic growth on a minimal medium ) . We conducted flux balance analysis of each model after standardizing model objective functions to the iFF708 biomass objective , and then applying constraints to the glucose , oxygen , and nitrogen exchange fluxes , first individually and then in combination . We used the measured uptake values reported by Österlund et al . [37] as constraints for each of these exchange reactions . | Scientists have been mapping the chemical reactions cells use to grow and manage waste since before enzymes were first identified more than 150 years ago . The model yeast Saccharomyces cerevisiae has one of the most extensively studied metabolic networks , including at least 25 metabolic network models published since 2003 . If iterative model improvement refines the metabolic network map , we would expect eventual convergence to a full , accurate metabolic network reconstruction . In this study , we looked for evidence of such convergence through comparative analysis of 12 genome-scale yeast models . We conducted simulations and evaluated model features such as predictive accuracy , genomic coverage and the included metabolites and reactions . We found that no single metric for evaluating models can adequately summarize important aspects of model quality . In some cases , we observed tradeoffs between model predictive accuracy and network coverage . We found evidence of incremental changes to the network reconstruction , but not marked shifts in model predictive ability or other metrics clearly arising from changes to the network alone . This work has broader implications to computational reconstruction of metabolic networks for any organism , and suggests that there is opportunity for refocusing the model building process to better support mapping cellular metabolic networks . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Comparative Analysis of Yeast Metabolic Network Models Highlights Progress, Opportunities for Metabolic Reconstruction |
The mechanism ( s ) by which bacterial communities impact susceptibility to infectious diseases , such as HIV , and maintain female genital tract ( FGT ) health are poorly understood . Evaluation of FGT bacteria has predominantly been limited to studies of species abundance , but not bacterial function . We therefore sought to examine the relationship of bacterial community composition and function with mucosal epithelial barrier health in the context of bacterial vaginosis ( BV ) using metaproteomic , metagenomic , and in vitro approaches . We found highly diverse bacterial communities dominated by Gardnerella vaginalis associated with host epithelial barrier disruption and enhanced immune activation , and low diversity communities dominated by Lactobacillus species that associated with lower Nugent scores , reduced pH , and expression of host mucosal proteins important for maintaining epithelial integrity . Importantly , proteomic signatures of disrupted epithelial integrity associated with G . vaginalis-dominated communities in the absence of clinical BV diagnosis . Because traditional clinical assessments did not capture this , it likely represents a larger underrepresented phenomenon in populations with high prevalence of G . vaginalis . We finally demonstrated that soluble products derived from G . vaginalis inhibited wound healing , while those derived from L . iners did not , providing insight into functional mechanisms by which FGT bacterial communities affect epithelial barrier integrity .
Mucosal surfaces exposed to the external environment contain distinct bacterial communities that exist in relationship with the host and can contribute to health and functioning . These bacterial communities have been linked to several human diseases and overall health [1] , and can vary between individuals , but also over time within the same person [2] . In the female genital tract ( FGT ) , colonization by Lactobacillus species and other lactate-producing bacteria helps to inhibit colonization by pathogenic bacteria [3] . However , colonization by more diverse communities of anaerobic bacteria , notably Gardnerella vaginalis , is common [4] , and often associated with the development of bacterial vaginosis ( BV ) [5] . BV is highly prevalent , affecting 4–58% of women globally; some areas , such as sub-Saharan Africa have rates as high as 55% [6] . BV is associated with significant health consequences , including pre-term birth , post-partum endometriosis , pelvic inflammatory disease , upper reproductive tract infections , and increased susceptibility to sexually transmitted infections ( STI’s ) [7 , 8] , with HIV being highly significant [9 , 10] . Indeed , a recent meta-analysis linked BV to a 60% increase in HIV acquisition rates [11] . However , while these relationships between microbial composition and vaginal health have been described epidemiologically , there is limited understanding about the mechanisms underlying the impact of bacterial dysbiosis on the vaginal mucosa . Maintenance of the mucosal barrier is critical for preventing invading microorganisms , including HIV , from penetrating into tissues and entering circulation [12] . Bacterial diversity in the FGT has been strongly associated with negative consequences for FGT mucosa . Highly diverse communities dominated by G . vaginalis and Prevotella are associated with upregulated expression of Toll-like Receptor ( TLR ) and NFkB pathways , leading to increased pro-inflammatory cytokine concentrations and activation of immune cells [13] . While it is widely appreciated that BV is associated with inflammation , the mechanism that elicits this inflammation or the bacterial proteins associated with inflammation remain unresolved [14] , which may partly explain the limited effectiveness of antimicrobial treatment for BV [15–17] . Bacterial metabolites including hydrogen peroxide , antimicrobial peptides , and acids that reduce the FGT pH have been proposed to have an important impact in sustaining mucosal health [3] . Furthermore , the integrity of mucosal epithelial surfaces has been shown to depend on bacterial community composition in other diseases [18] , and has been proposed to be important in the FGT during bacterial dysbiosis [19] , but this has not been extensively studied . Each of these factors likely impact disease susceptibility independently , and a Lactobacillus-dominant microbiota likely contributes to many of these factors to maintain the function of the healthy FGT and inhibit infections . Taken together these studies suggest that host-microbe interactions are key to understanding negative consequences on vaginal health , yet this interaction remains poorly defined in human cohorts [20] . We sought to better understand the relationship between mucosal health and bacterial diversity using a combination of metaproteomics and metagenomics , which to our knowledge represents the first attempt at integrating these approaches to study the FGT . Indeed the functional diversity of the bacterial proteome , and how this relates to FGT health and inflammation has not been assessed comprehensively , and has largely been limited to 16S rRNA gene sequencing . Thus , we hypothesized that bacterial protein factors can influence FGT mucosal health and affect disease susceptibility . Here we characterized FGT bacterial communities in two distinct human cohorts , longitudinally and cross-sectionally , in asymptomatic and symptomatic women with BV , uncovering bacterial-host interactions leading to wound healing impairment .
Cervicovaginal secretion samples from two cohorts of women were evaluated to understand the mucosal environment associated with bacterial dysbiosis . We first assessed mucosal changes in women at BV+ or BV- time points ( Cohort 1 , n = 10 ) , through a combination of mass spectrometry ( MS ) and 16S rRNA gene sequencing . MS analysis identified 1123 unique proteins , including 434 human and 689 bacterial proteins from 64 species . To assess the diversity of the bacterial proteome , we quantified the relative proteome load of each bacterial genus in each sample by summing the total number of protein spectral counts assigned to each genus , an approach previously shown to directly correlate with colony-forming units [21] . We clustered the bacterial proteomes from the twenty samples using unsupervised hierarchical clustering . Two major bacterial proteomes were identified , dominated by either Lactobacillus iners ( Group 1 , or G1 ) or Gardnerella vaginalis ( Group 2 , or G2 ) ( Fig 1A , species-level taxonomy shown in S1A Fig ) , which were used for downstream comparisons . In G1 , L . iners proteins accounted for 87–100% of the total protein load while in G2 , G . vaginalis proteins accounted for 48–96% . Compared to those in G1 , the bacterial proteomes in G2 displayed significantly higher species diversity ( S1C Fig ) . G2 profiles also had higher overall bacterial protein load when normalized to the total protein content ( 0 . 34 log10 , +2 . 2 fold higher; S1E Fig ) . L . iners dominated the FGT bacterial proteome of eight of the 10 patients from Cohort 1 at the time point without clinically diagnosed BV , but not the remaining two . Patient “10“ , at the time point without BV , displayed a high abundance of G . vaginalis with a lower abundance of L . iners , and Patient “6”at the time point without BV had high levels of Lactococcus lactis and Streptococcus mitis . In contrast , all samples taken during episodes of BV had high abundances of proteins from G . vaginalis , Prevotella spp . , Streptococcus mitis , Escherichia coli , and Atopobium vaginae , which have been previously identified as part of BV-associated bacterial communities [7] . Bacterial community composition for G1 and G2 was confirmed by 16S rRNA gene sequencing ( Fig 1B ) . According to 16S rRNA gene sequencing , G1 communities were dominated by Lactobacillus spp . ( 26–99% of the total community ) , and G2 communities were dominated by Gardnerella , but at lower proportions than were detected by MS ( 17–66% of the total community ) . As with MS , bacterial genera detected in BV-positive individuals by 16S rRNA gene sequencing included Sneathia , Prevotella , Atopobium , Megasphera , and others . 16S rRNA gene sequencing also detected greater bacterial diversity in the G2 samples compared to G1 ( S1C Fig ) . Several species detected by 16S and not by MS included Leptotrichia , Fastidiosipila , Shuttleworthia , and Aerococcus . Overall , this demonstrates significant heterogeneity in the structure of FGT bacterial communities between clinically defined BV and asymptomatic time points , that G . vaginalis and other anaerobes associate with BV , and that specific species dominate the bacterial proteome landscape in mucosal secretions . Mucosal samples from a separate group of 31 women from North America ( Cohort 2 ) were analyzed to further evaluate associations between FGT bacterial proteome diversity and BV . MS analysis of Cohort 2 samples showed similar trends to that of Cohort 1 ( Fig 1C , species-level taxonomy shown in S1B Fig ) , including Lactobacillus spp . -dominant ( G1 ) and G . vaginalis-dominant ( G2 ) communities . A wider distribution of lactobacilli including L . iners , L . crispatus , and L . jensenii was observed in G1 in Cohort 2 than Cohort 1 . Varying abundances of other BV-associated bacteria including Prevotella spp . , Atopobium vaginae , Mobiliuncus mulieris , and Sneathia sp . were also observed . In Cohort 2 , there was no difference in the species diversity of G1 compared to G2 ( S1D Fig ) . While all 7 women with BV clustered into G2 , 46% of participants demonstrated a G . vaginalis-dominated proteome despite a lack of clinical BV diagnosis , consistent with the observation that not all women have Lactobacillus-dominant FGT microflora despite low Nugent scores . Also similar to Cohort 1 , G2 in Cohort 2 had higher overall microbial proteome burden than G1 ( 1 . 5-fold higher; S1F Fig ) , indicating further changes in bacterial community function . This agrees with other studies showing Lactobacillus dominance varies between 37–90% of women , with greater diversity and variation in African women [5 , 13 , 22] . As bacterial diversity has been associated with other biological variables , such as concurrent STI’s [23] and hormonal contraceptive usage [10] , we compared clinical characteristics between Lactobacillus and Gardnerella-dominant groups ( Cohort 1-Table 1; Cohort 2-Table 2 ) . With the exception of BV status , we found no differences between G1 and G2 with respect to age , contraceptive use , antimicrobial usage , last menstrual period , detectible STI’s , or sexual practices in either cohort . There were differences in Amsel’s criteria collected from Cohort 2 ( S1 Table ) between G1 and G2 , where vaginal pH , clue and white blood cell presence was higher in women with G2 bacterial proteome profiles , in agreement with clinical BV status . Overall , there was no evidence to support that vaginal bacterial profiles were related to exogenous hormonal contraceptive use , the menstrual cycle , sexual behaviors , or concurrent STI’s in these cohorts . As the functional diversity of FGT-resident bacteria remained undefined we characterized the major bacterial pathways present in G1 and G2 profiles ( Fig 2 ) . Major functional categories represented in either group included transport and catabolism ( G1 average 19 . 5%; G2 average 12 . 6% ) , carbohydrate metabolism ( 1 average 14 . 5%; G2 average 15 . 7% ) , as well as nucleotide and amino acid metabolism ( G1 average 2 . 6%/1 . 8%; G2 average 2 . 0%/1 . 4% ) . However , unique functional signatures were observed between G1 and G2 FGT bacterial communities . Across both cohorts , the G1 group showed significant enrichment of proteins involved in transport and catabolism ( 6 . 9% higher ) , energy metabolism ( 5 . 6% higher ) , and folding , sorting , and degradation ( 4 . 9% higher ) , while G2 was highly significantly enriched in membrane transport functions ( 22% higher ) . Twelve bacterial proteins were significantly differentially abundant after multiple comparison correction between G1 and G2 in Cohort 1 ( S2 Table ) . Proteins enriched in G1 mostly belonged to L . iners proteins and were involved in homolactic fermentation of carbohydrates including glyceraldehyde-3-phosphate dehydrogenase ( GAP-DH ) , pyruvate kinase ( PK ) , and lactate dehydrogenase ( LDH ) . Proteins enriched in G2 were all G . vaginalis proteins and included a MalE-type ABC sugar transport system periplasmic component ( MAL-E ABC ) and an alpha-1 , 4 glucan phosphorylase , an enzyme that degrades starch and glycogen , suggesting that G . vaginalis directs its metabolism towards liberation and uptake of extracellular saccharides . Although many of these proteins were also differentially abundant between G1 and G2 in Cohort 2 they did not pass multiple comparison correction . Overall this shows that ‘core’ functional pathways necessary to host-associated bacterial life within the FGT include carbohydrate , amino acid , and translational machinery , while perturbations to membrane transport and carbohydrate catabolism are likely important for pathogenic states . Bacterial dysbiosis impacts HIV acquisition risk [10 , 11 , 24] , reproductive health [7] , and mucosal cellular activation [13] , but the effect on the FGT is not well defined . Our analysis revealed that 69/434 ( 15 . 8% , 15 passing 5% FDR ) and 64/434 ( 14 . 7% , 19 passing 5% FDR ) host proteins were significantly differentially abundant between G1 and G2 profiles in Cohort 1 and 2 , respectively . For Cohort 2 , comparison based on bacterial groups rather than Nugent score criteria yielded greater host proteome differences , statistically ( 9 . 2% vs . 15 . 8% , P<0 . 05 ) , and in magnitude ( 5 vs . 6 Log2 Fold Change; S1G/S1H Fig ) , suggesting that bacterial community composition , rather than clinical BV criteria , more accurately classifies mucosal inflammation . This comparison was not possible for Cohort 1 , as all G2 profiles had clinically defined BV . Hierarchical cluster analysis revealed that longitudinal changes from G1 to G2 profiles in Cohort 1 were clearly distinguishable by two major branches of host proteins ( S2A/S2B Fig ) . Proteins more abundant in G1 ( Branch 1 ) associated with epidermis development and the cornified envelope , whereas G2 ( Branch 2 ) showed increased factors involved in cytoskeletal-binding , threonine proteases involved in proteasome activity , as well as vesicular components and the melanosome . Many of these included S100 proteins and innate immune factors , important for antimicrobial defense based on gene ontology ( DMBT1 , CADH1 , S10A7 , EFHD2 , S10AB , S10A6 , TGM3 , K2C1 S10A2 ) . Similarly , in Cohort 2 , hierarchical cluster analysis showed that proteins more abundant in G1 ( Branch 1 ) associated with epidermis development , structural molecular activity , and the cornified envelope , while proteins elevated in G2 ( Branch 2 ) also included ectoderm development and differentiation , although were related to cytoskeletal activity ( S2C/S2D Fig ) . Many of these are important for leukocyte-mediated immunity and wounding responses based on their gene ontology ( A1AT , IC1 , GELS , CO3 , PEBP1 , PRDX1 , PRDX2 , CO4A , ANXA8 ) . Seventeen proteins were differentially abundant across both cohorts ( Fig 3A ) . Host proteins more abundant in G2 profiles included apoptotic regulators ( PRDX , NDKB , CADH1 ) and leukocyte migration factors ( PLST ) , while G1 profiles showed increased keratinization , epidermis development , and cornified envelope ( INVO , SPR1A ) factors ( Fig 3B ) . Of particular interest , the abundances of INVO and SPR1A were 14 . 7 and 7 . 2-fold lower in women G2 microbial profiles Cohorts 1 and 2 , respectively . In Cohort 2 , INVO and SPR1A were lower for women with G2 microbial profiles even if they had not been clinically diagnosed for BV ( Fig 3C ) . These proteins are known to act as scaffolding for epidermal layers and are important for proper barrier function [25] , and immunohistochemical analysis confirmed the presence of INVO and SPR1A in cervical and vaginal tissues , where they strongly associated with the squamous epithelium and stratum corneum in healthy FGT tissue ( S3 Fig ) . Collectively these data show an association of heightened immune activation , apoptosis , and decreased epithelial barrier function in women with G . vaginalis-dominated bacterial profiles and that these effects are evident in G . vaginalis-dominated communities in the absence of clinical diagnosis . Due to the strong association of epithelial development pathways with different bacterial groups , we compared cornified envelope factors INVO and SPR1A to bacterial proteins . Nineteen bacterial proteins had strong associations in at least one comparison against either INVO or SPR1A after correcting for multiple comparisons . Proteins from L . iners that positively correlated with INVO and SPR1A were involved in Catabolism and Energy Metabolism pathways , including glycolysis and homolactic fermentation of sugars ( Embden-Meyerhoff-Parnas ( EMP ) pathway ) ( Fig 4A/4B ) . These included a putative fructose 1 , 6-bisphosphate aldolase ( PFBA ) , PK , GAP-DH , and LDH , as well as a ferritin-like protein ( FLP ) , which is important for sequestering excess iron and preventing oxidative damage [26] . Bacterial proteins that negatively correlated with INVO and SPR1A belonged to alternate sugar metabolism pathways ( phosphoketolase pathway ) , transport functions , and amino acid catabolism . The majority of these belonged to G . vaginalis ( Fig 4C/4D ) , including D-xylulose 5-phosphate/D-fructose 6-phosphate phosphoketolase ( XFBP ) , a putative sugar-binding secreted protein ( P-SBSP ) , MAL-E ABC , an extracellular solute binding protein ( ESBP ) , and glycine oxidase ( GOx ) . A membrane protein from Prevotella sp . , sharing sequence homology with SusD-like ( starch-binding ) protein , was also negatively associated ( S4 Fig ) . Many associations with vaginal pH were also observed , including negative associations with enzymes from L . iners ( PK , LDH , elongation factor tu , and FLP ) ( S5A Fig ) , and positive associations with enzymes from G . vaginalis , including GOx , which catalyzes the conversion of glycine into glyoxalate , ammonia , and hydrogen peroxide ( S5B Fig ) . Therefore a clear relationship between metabolic function , epithelial barrier protein levels , and vaginal pH was observed , demonstrating that these microbial pathways may be an important component of mucosal barrier disruption and vaginal health . The association of bacterial communities with barrier integrity proteins led us to hypothesize that wound-healing capacity may be supported or inhibited by specific bacterial species and/or their products . We thus performed a classical wound-healing assay wherein we cultivated relevant cervical cell line ( HeLa CCL-2 ) in the presence of supernatants derived from cultures of L . iners or G . vaginalis . Prior to adding culture supernatants , a wound was induced by scratching HeLa cell monolayers . Incubation of scratched monolayers with L . iners culture supernatant did not alter wound healing compared to the control incubations . However , incubation with G . vaginalis culture supernatants significantly reduced wound healing after 24 hours compared to both the control and L . iners conditions ( Fig 5 ) . These results confirm a relationship between soluble compounds produced by the major bacterial species of the G1 and G2 profiles and wound healing capacity . This implicates these species as important components or drivers of epithelial barrier repair , maintenance , and disruption in the FGT .
In this study , we demonstrated a novel metaproteomic approach to simultaneously assess bacterial diversity , abundance , and function , along with host barrier and inflammation processes , providing mechanistic insight relevant to women’s health . We described distinct vaginal bacterial proteome profiles that were dominated by Lactobacillus spp . ( G1 ) or G . vaginalis ( G2 ) , where the latter associated with BV , increased community diversity , and significant divergence from normal metabolic function . We next demonstrated that bacterial functional profiles were significantly associated with cornified envelope factors in the FGT , and this was affected even in the absence of clinical diagnosis . Finally , we found that predominant species identified in this study , specifically G . vaginalis and L . iners , generate soluble products that disrupt or maintain the ability of cervical epithelial cells to repair and close wounds . Therefore , impaired wound healing is a potential mechanism by which key bacterial species may impact mucosal barrier function and therefore disease and/or HIV/STI infection risk . The association of vaginal inflammation and inflammatory vaginal bacteria with HIV susceptibility indicates that targeting this mechanism may lead to novel prevention strategies for HIV . While increased diversity of bacterial communities has been linked to better mucosal functioning in the gut [27 , 28] , low-diversity bacterial communities are beneficial for the FGT [5] , where increased diversity is strongly associated with BV [7] . Consistent with previous observations [5] , many of the women in Cohort 2 with G . vaginalis-dominated communities were asymptomatic for BV ( 61% ) , further supporting the fact that Nugent score is underestimating the extent of non-Lactobacillus dominant communities . However , the effects on host epithelial pathways , including decreased integrity and increased inflammatory pathways were still evident in the absence of clinical diagnosis . Lactobacillus spp . and G . vaginalis proteins comprised more of the soluble proteome load than might be inferred from 16S rRNA gene sequencing , suggesting that these bacteria dominate the metabolic landscape of the FGT . Metagenomic studies of the human microbiome have shown that core metabolic function is less variable than the community composition [2] . In agreement with this , we observed that the majority of assigned protein functions did not vary significantly , which likely represent core metabolic functions . However , some functions varied between G1 and G2 , including increased carbohydrate metabolism , energy production , and folding/sorting functions in G1 to enhanced membrane transport and secretion of extracellular products in G2 , with L . iners and G . vaginalis dominating these key functions . The increased abundances of enzymes important for sugar transport and starch and glycogen catabolism in G2 suggest that G . vaginalis may outcompete Lactobacillus spp . for the uptake of carbohydrate substrates . This agrees with a recent study showing that women with BV have significant metabolite alterations in cervicovaginal mucous , including lower levels of carbohydrates , amino acids , and lactate , accompanied by increased levels of amino acid catabolites and polyamines [29] . Overall , this demonstrates that increased bacterial diversity is associated with changes in key metabolic pathways , which allows for better understanding of dysbiosis in the FGT . We found that G1 profiles from both cohorts strongly associated with cornified envelope factors , especially INVO and SPR1A , which are expressed in the upper layers of the vaginal and cervical epithelia , and aid in maintaining epithelial integrity . Our group has previously reported that increased levels of cervocovaginal CD4+ T cells associated with lower levels of cornified envelope factors [30] , demonstrating the important link between vaginal epithelial integrity and HIV acquisition risk . G1 profiles were also associated with higher levels of antimicrobial peptides , such as dermcidin , which is important for host defense against microorganisms [31] . In comparison , the G2 bacterial profiles correlated with lower cornified envelope and epithelial barrier factors , increased cytoskeletal elements important for cell migration , and increased proteasome factors . This agrees with other studies showing that BV associates with activation of innate immune and inflammation pathways in the FGT , including increased complement [32] , proteasome levels [33] , and pro-inflammatory cytokines and activated CD4+ T-cells [13] . Importantly , G2 bacterial proteome profiles associated with decreased abundances of INVO and SPR1A regardless of clinical BV status . This finding demonstrates that current methods used to diagnose BV likely underestimate the true extent of bacterial dysbiosis on mucosal barrier function the FGT , as the Nugent Scores were poor predictors of BV , especially for Cohort 2 . Thus , new methods to detect and treat G . vaginalis in the FGT could aid in reducing HIV acquisition risk by promoting mucosal and epithelial barrier integrity , and reduced inflammation . Catabolic enzymes involved in homolactic fermentation of glucose from Lactobacillus , such as L-lactate dehydrogenase , correlated with higher epithelial barrier proteins , while membrane transporters , extracellular proteins , and alternate routes of carbohydrate metabolism ( heterolactic fermentative or phosphoketolase pathways ) from G . vaginalis were negatively correlated . In addition , GOx , was strongly correlated with increased vaginal pH , implicating a role of this enzyme in altered vaginal pH during dysbiosis . To our knowledge this is the first time these bacterial enzymes have been associated with epithelial disruption signatures and vaginal pH . Collectively , this shows a relationship between bacterial community structure , metabolic function , disruption of epithelial proteins important for barrier integrity , and overall vaginal health . We also demonstrated that G . vaginalis culture supernatants inhibited healing of scratched HeLa cell monolayers while , L . iners culture supernatants maintained effective wound healing . Based on these data , G . vaginalis is likely an important component or a potential driver of subverting the wound healing process . While acknowledging that HeLa cell monolayers do not completely recapitulate the squamous epithelium or immune environment of the FGT , this nevertheless supports would healing as an underlying mechanism . Taken collectively , and considering the metaproteomic , metagenomic , and in vitro models , these data suggest that G . vaginalis releases a variety of extracellular products in the vaginal compartment that aid in uptake for nutrients , alter the vaginal microenvironment , contribute to innate immune activation , and prevent healing of the epithelial barrier . Future studies to identify exact protein pathways involved , how they may be altered , and more advance animal and engineered tissue models would help better decipher these host-microbiome interactions . It is important to compare discuss the benefits and limitations of metaproteomics compared to 16S rRNA-based techniques to characterize microbial communities in the vaginal compartment . Both techniques are quantitative and spectral counts by MS have been shown to correlate directly to colony-forming units [21] . An advantage of 16S over MS is greater resolution of the overall community structure , and while we showed high sensitivity to identify species that were at 0 . 1% of the population by MS , 16S captured more overall bacterial species . It is likely that the larger dynamic range of the proteome over the genome is a large contributing factor to this observation . Both 16S and MS rely on curated databases to identify species and are subjected to this same limitation in availability and extensiveness of libraries . While databases for 16S rRNA genes are likely more comprehensive , proteomic libraries are growing and becoming more available . MS is advantageous in that it can provide direct species-level identification , which is not achievable through high-throughput 16S rRNA gene sequencing methods . Furthermore , metaproteomic analysis reveals bacterial functional and metabolic activity , which is not provided by 16S-based approaches . Prior studies have attempted to alleviate this using MS to correlate metabolite abundances with species abundances [34] , through metagenomic studies [35] , or by employing computational methods to estimate bacterial community functional capacity based on 16S rRNA gene signatures [36] , but nevertheless represent indirect methods to evaluate bacterial community functionality . While 16S rRNA gene sequencing is a popular and well-validated method for studying microbial communities , the use of metaproteomic approaches provides complimentary and invaluable data on community structure , function , and host inflammation to better study host-bacterial relationships . Our data provide novel mechanistic insight of how dysbiosis of vaginal bacterial communities may directly increase host susceptibility to infection through the disruption of epithelial barriers , inhibition of wound repair , and induction of inflammation . In the context of HIV transmission , inhibition of wound repair is under studied and may represent underlying mechanisms in other risk factors for HIV , including hormonal contraceptive usage , intravaginal practices , and other STI’s . These pathways may also impact the effectiveness or responsiveness to mucosa-targeted prevention technologies for other infections , such as microbicides or vaccines for HIV . In summary , this study delineated functional configurations of microbial communities that impact vaginal health during BV , providing new information on host-bacterial interactions , enabling future experiments to probe host-microbe relationships in the FGT that could have important implications for women’s health .
All women who participated in this study provided written informed consent . The studies were approved by the University of Washington Human Subjects Review Committee , the Kenya Medical Research Institute ( KEMRI ) , Human Subjects Committee of the University of Illinois at Chicago , and the Research Ethics Board of the University of Manitoba . Vaginal swabs were eluted with 2 x 250ul washes in PBS ( pH 7 . 0 ) . Swab eluates ( Cohort 1 ) or CVL samples ( Cohort 2 ) were then centrifuged in SpinX tubes with a bonded fritted bottom ( Corning , Corning , NY ) , and protein content determined by BCA assay ( Novagen , Bilerica , MA ) . Proteins were then denatured , reduced , alkylated , digested into peptides , and prepared for mass spectrometry as described previously [38] . Detailed methods for this process are available in S1 Methods . Briefly , peptide samples were injected into a nano-flow LC system ( Easy nLC , Thermo Fisher ) connected inline to a LTQ Orbitrap Velos ( Thermo Fisher ) mass spectrometer , and analyzed in a label-free manner as described previously [38] . Peptide identity searching was performed with Mascot v2 . 4 . 0 ( Matrix Science ) against a manually curated database comprised of the SwissProt Human & Bacteria ( June 2015 ) and UniProtKB/Trembl All Bacteria databases ( August 2015 ) . A decoy database was included to determine the rate of false discovery . Protein identifications were confirmed using Scaffold ( v 4 . 4 . 1 , Proteome Software ) with confidence thresholds set at 95% protein identification confidence , requiring at least 2 unique peptides and 80% peptide identification confidence . A combination of label-free methods was used for protein quantitation: spectral counting ( for microbial proteins and bacterial diversity clustering , see below ) and area-under-the-curve quantitation ( Progenesis LC-MS software ( v4 . 0 , Nonlinear Dynamics ) ) . Criteria for assigning presence of microbial proteins included those that had at least 1 peptide in one sample , and at least 2 peptides per protein across all samples . These parameters resulted in a false discovery rate below 3 . 1% based on the search results run against Mascot’s generated decoy database . For the latter , only proteins that had an average co-variance of <25% ( 575 proteins ) , as determined through measurements of standard reference sample run at 10 sample intervals ( total 7 times ) , were utilized in downstream analysis to exclude proteins with higher technical measurement variability . Complete details of liquid chromatography and mass spectrometry instrument settings are as described previously [38] . Biological/molecular functions and cellular components were annotated based on gene ontologies using the DAVID Bioinformatics Resource ( v6 . 7 ) [41] , which calculates a modified Fisher’s Exact P value to determine the probability that the association between each protein in the dataset and functional pathway is random . Functional categories were considered to be those with P-values < 0 . 05 ( Benjamini Hochberg adjusted ) and at least 3 proteins selected to be positive associations . HeLa ( ATCC CCL-2 ) cells were obtained as a gift from the laboratory of Dr . Shiu-Lok Hu ( University of Washington ) , and were maintained in Dulbecco’s Modified Eagle’s Medium ( DMEM ) supplemented with 4 . 5 g/L glucose , L-glutamine , 10% ( v/v ) fetal bovine serum ( Corning ) , and 1% ( v/v ) penicillin/streptomycin/amphotericin B solution ( Gibco ) . HeLa cells were incubated at 37°C with air/5% CO2 atmosphere . Gardnerella vaginalis ATCC 14018 and Lactobacillus iners ATCC 55195 were obtained from the American Type Culture Collection , and were maintained using HBT-Bilayer medium ( BD ) and NYCIII liquid medium with incubation at 37°C with air/5% CO2 atmosphere . Frozen stocks were stored in 20% ( v/v ) glycerol at -80°C . To assess the impact of different bacteria on the ability of cervical epithelial cells to repair wounds , we utilized the well-established in vitro scratch assay [42] . To prepare live bacteria and culture supernatants for the wound-healing assay , overnight cultures of L . iners and G . vaginalis in NYCIII medium were grown as described above . Wells of a 24-well tissue culture plate ( Corning ) were initially seeded with 50 , 000 HeLa cells in a volume of 500 μL DMEM and incubated at 37°C under 5% CO2 until a confluent cell monolayer had formed . Monolayers in each well were then scratched using a sterile P200 pipette tip . Live bacteria , bacterial culture supernatants , or control solutions were then added to the wells . Images at five reference points per well were captured using a Nikon Eclipse TS100 microscope equipped with a Nikon DS-Ri1 camera and the size of the scratch at each reference point was manually analyzed using the ImageJ software . The size of the wound was determined immediately after beginning ( t = 0 ) the experiment and then again after 24 hours ( t = 24 ) of incubation at 37°C with air/5% CO2 atmosphere . Additional information on wound-healing assays is available in S1 Methods . | The female genital tract ( FGT ) is a key mucosal surface in the context of HIV transmission . Lactobacillus species are beneficial to the FGT , while Garderella vaginalis and other anaerobic bacteria are detrimental . Bacterial vaginosis ( BV ) is an inflammatory condition characterized by an outgrowth of G . vaginalis and other anaerobes , which is linked to increased HIV acquisition rates . However , the mechanism behind this remains unknown . Here , we used a novel proteomic approach to simultaneously evaluate host and bacterial functions in the FGT . We found that women with G . vaginalis-dominated FGT bacterial communities always displayed markers of decreased epithelial barrier integrity , and decreased wound healing capacity . We also demonstrated that the abundance of proteins from G . vaginalis associated with these signatures of disrupted epithelial integrity . Finally , we showed that products derived from G . vaginalis prevented healing of wounded cell monolayers while products derived from L . iners maintained the ability of the cell monolayers to close wounds . This study provides novel mechanistic insight into the link between BV and increased HIV acquisition rates . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"sequencing",
"techniques",
"urology",
"medicine",
"and",
"health",
"sciences",
"gut",
"bacteria",
"bacterial",
"vaginosis",
"physiological",
"processes",
"gene",
"sequencing",
"sexually",
"transmitted",
"diseases",
"molecular",
"biology",
"techniques",
"cellular",
"structures",
"and",
"organelles",
"bacteria",
"wound",
"healing",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"tissue",
"repair",
"proteins",
"structural",
"proteins",
"molecular",
"biology",
"lactobacillus",
"ribosomes",
"biochemistry",
"rna",
"ribosomal",
"rna",
"cell",
"biology",
"nucleic",
"acids",
"proteomes",
"genitourinary",
"infections",
"physiology",
"biology",
"and",
"life",
"sciences",
"non-coding",
"rna",
"dna",
"sequencing",
"organisms"
] | 2016 | Microbiome Composition and Function Drives Wound-Healing Impairment in the Female Genital Tract |
One-third of the 4 , 225 protein-coding genes of Escherichia coli K-12 remain functionally unannotated ( orphans ) . Many map to distant clades such as Archaea , suggesting involvement in basic prokaryotic traits , whereas others appear restricted to E . coli , including pathogenic strains . To elucidate the orphans' biological roles , we performed an extensive proteomic survey using affinity-tagged E . coli strains and generated comprehensive genomic context inferences to derive a high-confidence compendium for virtually the entire proteome consisting of 5 , 993 putative physical interactions and 74 , 776 putative functional associations , most of which are novel . Clustering of the respective probabilistic networks revealed putative orphan membership in discrete multiprotein complexes and functional modules together with annotated gene products , whereas a machine-learning strategy based on network integration implicated the orphans in specific biological processes . We provide additional experimental evidence supporting orphan participation in protein synthesis , amino acid metabolism , biofilm formation , motility , and assembly of the bacterial cell envelope . This resource provides a “systems-wide” functional blueprint of a model microbe , with insights into the biological and evolutionary significance of previously uncharacterized proteins .
Because of its central position in the microbial research community , the Gram-negative bacterium Escherichia coli plays a leading role in investigations of the fundamental molecular biology of bacteria [1–8] . This experimentally tractable microbe is a workhorse in basic and applied research aimed at elucidating the mechanistic basis of prokaryotic processes and traits , including those of pathogens . The ever-expanding availability of genomic resources makes E . coli particularly well-suited to systematic investigations of microbial protein components and functional relationships on a global scale . These include a genome-wide collection of single-gene deletion strains [2] along with extensive knowledge of regulatory circuits [3 , 5 , 7 , 9] and metabolic pathways [6 , 10 , 11] . Yet despite being the most highly studied model bacterium , a recent comprehensive community annotation effort for the fully sequenced reference K-12 laboratory strains [8] indicated that only half ( ∼54% ) of the protein-coding gene products of E . coli currently have experimental evidence indicative of a biological role . The remaining genes have either only generic , homology-derived functional attributes ( e . g . , “predicted DNA-binding” ) or no discernable physiological significance . Some of these functional “orphans” ( not to be confused with ORFans , which are genes present within only single or closely related species ) may have eluded characterization in part because they exhibit mild mutant phenotypes , are expressed at low or undetectable levels , or have limited homology to annotated genes . This suggests more-sensitive analytical procedures are warranted . A key feature of the molecular organization of all organisms , including bacteria , is the tendency of gene products to associate into macromolecular complexes , biochemical pathways , and functional modules that in turn mediate all the major cellular processes . Elaboration of these interaction networks via proteomic , genomic , and bioinformatic approaches can reveal previously overlooked components and unanticipated functional associations [12] . For example , a recent integrative analysis of phenotypic , phylogenetic , and physical interaction data led to the discovery of an evolutionarily conserved set of novel bacterial motility-related proteins [13] . However , although systematic integration of diverse high-throughput interaction datasets is routinely performed to reveal new functional relationships in model eukaryotes such as yeast , worm , and fly [14–19] , few analogous studies of the global functional architecture of E . coli , and any prokaryote for that matter , have been reported to date [20–22] . To this end , we have combined complementary , highly sensitive computational and experimental procedures to derive extensive high-quality maps of the functional interactions inferred by genomic context ( GC ) methods and physical interactions ( PI ) deduced by proteomics of E . coli . Our results indicate that many previously unannotated bacterial proteins are components of functionally cohesive modules and multiprotein complexes linked to well-known biological processes . A substantive fraction of these associations could be verified by independent experimentation and were found to be broadly conserved across prokaryotic phyla , indicating homologous systems in other microbes , whereas others are seemingly restricted to the E . coli lineage . The entire data collection is publicly accessible via a searchable Web-browser interface to stimulate exploration of both conserved and specialized bacterial proteins within the context of biological processes of particular interest .
Since the functional characterization of E . coli , and bacteria in general , has largely been guided historically by scientific interests and technical considerations , some bias is expected in terms of the coverage and depth of existing biological knowledge as reflected in current gene annotations . To evaluate the degree to which the physiological functions of the 4 , 225 putative protein-coding sequences of E . coli K-12 are characterized presently , we examined the scope of literature reference records curated in the UniProt annotation system [23] . After excluding PubMed references corresponding to genomic mapping studies , the average total number of papers associated with each of the proteins of E . coli K-12 is surprisingly limited ( Figure 1A ) , with many proteins apparently still uncited . We next examined recent E . coli K-12 ( substrains W3110 and MG1655 ) gene annotations in the public databases RefSeq [24] , MultiFun [25] , and EcoCyc [11] . Since W3110 is commonly used for high-throughput studies , we devoted the bulk of our subsequent analysis to this substrain . However , to make sure that relevant functional attributes were not overlooked , we cross-mapped the corresponding gene accessions in both substrains and compiled an inclusive set of functional annotations accordingly ( Table S1 ) . In total , we found that 2 , 794 ( 66% ) of E . coli's proteins had either proper mnemonic names [26] , experimentally derived annotations in the MultiFun multifunction schema , or literature documentation to a well-defined pathway or multiprotein complex in EcoCyc ( Figure 1B ) . This left 1 , 431 proteins ( 34% ) as currently functionally uncharacterized ( which constitute our orphans set , listed in Table S1; see Protocol S1 for details ) . Of these , 446 ( 31% ) have at least one putative molecular function defined on the basis of sequence ( such as the presence of a predicted DNA-binding domain or an enzymatic motif ) in the Clusters of Orthologous Groups ( COGs ) of proteins catalog [27] . The genes lacking annotation appear to be translated into bona fide proteins as their corresponding transcripts [28] were not significantly ( p = 0 . 36 ) less stable than the products of annotated genes ( Figure 1C ) . However , some differences were evident in terms of their biophysical attributes and evolutionary scope relative to annotated genes ( Table S1 ) . Most notably , only 21 orphans ( 1 . 5% ) are required for viability under standard laboratory conditions [2] in contrast with the 280 annotated genes ( 10% ) previously deemed essential . The orphans were also significantly ( p < 1e−10 ) less abundant at both the transcript ( Figure 1D; average normalized mRNA expression over 400 microarray experiments [5]: 8 . 0 [orphans] vs . 8 . 9 [annotated] ) and protein levels ( Figure 1E; average codon adaptation index [CAI]: 0 . 41 vs . 0 . 47 ) . Furthermore , they tend to encode somewhat smaller proteins ( Figure 1F; average MW: 29 . 4 vs . 38 . 2 kDa; p < 1e−10 ) with fewer domain assignments ( 44% ) than for annotated proteins ( 74% ) according to the SUPERFAMILY database [29] ( Table S1 ) . Orphans also generally find fewer orthologs in a nonredundant genome dataset , filtered at 90% similarity based on the frequency of shared orthologs ( Figure 1G ) , with an average of 0 . 22 as compared with 0 . 48 for annotated genes ( p < 1e−10 ) using a maximum-score E-value cutoff of 1 × 10−6 for BLAST bidirectional best hits ( BDBHs; Table S2 and Protocol S5 for details ) . Nevertheless , broader sequence comparisons against currently available metagenomes ( Figure 1H ) indicated that orphan homologs ( one-way BLAST hits ) are often widely distributed in diverse environments ( Table S2 and Protocol S2 ) ; for example , a high proportion ( 0 . 80 ) of orphans have homologs present in marine metagenomes , anaerobic bacterial populations ( farm silage , 0 . 51; whalefall , 0 . 50; sludges , 0 . 49 ) , and even in the residents of the mammalian gut ( union of human and mouse , 0 . 35 ) , implying participation in core bacterial processes . Furthermore , the same high proportion ( ∼99% ) of orphan and annotated genes have orthologs in the other sequenced E . coli isolates , including pathogenic variants and closely related Shigella strains ( Table S2 ) . Taken together , this argues that the functional significance of the orphans is more pervasive than the current annotations suggest . The scarcity of the existing knowledge regarding the biological roles of the orphans is likely due to multiple reasons , ranging from the lower expression , nonessentiality , or smaller sizes of certain orphan proteins to their lack of obvious homologs in other organisms including humans . Accordingly , integration of multiple data sources is warranted to decipher the specific biological roles of this uncharacterized repertory . Since the elucidation of physical and functional interaction networks can provide insights into bacterial protein function based on the concept of guilt by association [30] , we took a multipronged approach . We performed large-scale proteomic analysis to determine orphan participation as components of multimeric protein complexes , and inferred functional relationships based on genomic context inference , which exploits the patterns of gene conservation across bacterial genomes [31] . We then predicted the functions of the orphans using an integrative machine-learning procedure with extensive benchmarking . Finally , we performed independent experiments to validate a subset of high-confidence predictions related to core biological processes . Key steps in our pipeline are outlined schematically in Figure 2 . We performed systematic large-scale tandem-affinity purifications of all endogenous soluble orphan and annotated proteins detectably expressed in E . coli W3110 under standard culture conditions ( see Materials and Methods and Protocol S3 for details ) . We used an optimized Sequential Peptide Affinity ( SPA ) -tagging system to isolate multiprotein complexes [32] . This procedure is based on the integration of a marker cassette bearing a dual-affinity tag , consisting of three FLAG sequences and a calmodulin-binding peptide separated by a protease cleavage site , fused to the C-termini of targeted open reading frames in E . coli DY330 ( W3110 background ) via λ-phage “Red”–mediated homologous recombination . This system enables recovery of native bacterial protein complexes at near-endogenous levels [4] , minimizing spurious nonspecific protein associations . Stably interacting polypeptides were subsequently detected using a highly sensitive combination of tandem mass spectrometry ( LCMS ) and peptide mass fingerprinting procedures ( MALDI ) to increase detection coverage and accuracy ( Protocol S3 ) , just as we had previously done in a focused investigation of highly conserved essential E . coli proteins [4] . We successfully chromosomally tagged 1 , 241 new baits , aiming to verify putative interactions by reciprocal tagging where possible , for a total of 1 , 476 large-scale protein purifications ( after including the 235 reported previously ) , of which 552 represented orphans ( Protocol S3 ) . Since proteomic datasets typically contain noise in the form of nonspecific associations , we performed a careful statistical analysis and quality filtering to determine biologically meaningful PI . We considered that the specificity and affinity between any two putatively interacting proteins should be correlated with the consistency of copurification over all the experiments in which the proteins were identified ( i . e . , co-complexed ) . We therefore used an established copurification metric [33] to assess interaction specificity based on the similarity of the protein copurification patterns ( Protocol S3 ) . We then generated a single consolidated confidence score for each putative pairwise physical interaction based on the copurification metric together with the primary interaction evidence to penalize inconsistent or promiscuous binders ( i . e . , possible false positives ) using alternatively a logistic regression model and Bayesian inference [34] ( Protocol S3 ) . The logistic regression model was trained using a reference set of curated gold-standard PI ( Protocol S3 ) , which represents the union of experimentally verified PIs derived from low-throughput experiments extracted from the Database of Interacting Proteins ( DIP ) [35] , the Biomolecular Interaction Network Database ( BIND ) [36] , and the IntAct database [37] . For the negative gold standards , we compiled pairs of proteins annotated with different subcellular localizations ( i . e . , one cytoplasmic , the other periplasmic or outer membrane-bound [38] . Despite its relative simplicity , the logistic regression model offered better performance than the Bayesian method ( see Figure 3A and Table S3 ) . We therefore applied the former to our global PI network , assigning a probabilistic confidence score for each pair of putatively interacting proteins ( Protocol S3 ) . To minimize false positives without incurring excessive false negatives , we further filtered our network using a stringent minimum confidence cutoff of ≥0 . 75 as a high proportion ( 71% ) of PI verified by reciprocal purification ( Table S4 ) had likelihood scores at or above this threshold ( Figure S1A ) . Finally , we removed the ten most-highly connected “hub” proteins that were deemed particularly abundant nonspecific contaminants ( Table S5 ) . The resulting final network consisted of 5 , 993 high-confidence , nonredundant pairwise interactions among 1 , 757 distinct E . coli proteins , including 451 orphans , or roughly two-thirds of the predicted soluble cytoplasmic proteome . As summarized in Figure 3B , most ( 3 , 193 , or 53% ) of these PI are novel ( Table S6 ) , whereas only 47% were already reported in either the DIP , BIND , or IntAct interaction databases , or previous large-scale proteomic studies [1 , 4] . Importantly , our filtered dataset had a comparable level of accuracy ( median confidence of 0 . 79 ) as for the much smaller set of 716 “validated” PI previously reported by our group [4] and a genome-scale dataset of 7 , 123 PI ( median confidence of 0 . 69 ) generated using an analogous affinity purification schema in yeast [39] . The reliability of our dataset was also evident by two additional independent criteria . First , the mRNA expression patterns of the putatively interacting proteins were nearly as highly correlated as those of PI determined by low-throughput experiments ( Figure 3C ) , even when these last are formed by presumably more abundant proteins ( Figures S1E and S1F ) . Second , despite the more limited evolutionary distribution of the orphans , the putatively interacting proteins exhibited an elevated degree of co-occurrence of the respective orthologs across other bacterial species , evident in the high mutual information of the corresponding phylogenetic profiles ( Protocol S5 ) , again comparable to that of interacting pairs derived from low-throughput experiments ( Figure 3D ) . Collectively , these results indicate that our physical interaction network is very likely to be informative about orphan protein function . Since macromolecular assemblies mediate biological function in cells , we partitioned our high-confidence physical interaction network using the Markov clustering algorithm ( MCL; see Materials and Methods and Protocol S4 ) to define orphan membership as subunits of discrete multiprotein complexes . MCL simulates random walks ( i . e . , flux ) to delimit highly connected subnetworks based on both the connectivity and the weight of the graph edges [40] . In this case , the weights reflect the integrated PI scores obtained by logistic regression ( Figure 2A ) . The higher the flux within in a region of the PI network , the more likely MCL will delimit the region as a cluster ( in this case , a putative multimeric protein complex ) . A recent comparative study [41] found that MCL is often superior to other clustering algorithms in identifying functionally related groupings in probabilistic molecular interaction graphs and is remarkably resilient to spurious graph perturbations ( e . g . , missing edges ) . We optimized the MCL parameters ( see Materials and Methods and Protocol S4 ) to partition the 5 , 993 PI network , generating a set of 443 putative multiprotein complexes ( Figure 3E ) , most of which consist of two to four polypeptides ( Table S7 ) . In agreement with previous reports [41] , alternative clustering algorithms comparable to MCL in terms of accuracy , such as the Restricted Neighborhood Search Cluster algorithm [42] , produced similar groupings ( unpublished data ) . Moreover , as was found in a proteomic survey of yeast multiprotein complexes [39] , both the subunit number and degree of connectivity of the MCL clusters followed a power-law distribution ( Figure S2B and S2C ) . More telling , 244 ( 55% ) of these E . coli multiprotein complexes contained at least one orphan as a putative subunit , with linkages suggestive of concerted biological functions ( Figure 3E ) . The complexes also showed a significant ( p < 0 . 001 ) enrichment in terms of functional homogeneity , compared with null random models ( Figure S2A ) , implying that both the annotated components and the associated orphans tend to participate in the same biological processes . For example , 25 orphans were detected as part of a large subnetwork of putative complexes involved in protein synthesis ( Figure 3F ) . These include the orphans YbcJ and YncE , which physically interacted with the pseudouridylate synthase RluB , the RNA helicases SrmB and DeaD , the exoribonucleases E ( Rne ) and R ( Rnr ) , and other components of the ribonucleolytic “degradosome” responsible for mRNA degradation , suggesting a probable role in RNA processing and/or turnover . Likewise , YfgB copurified with three translation-related complexes , including ribosomal proteins . Consistent with these observations , the expression of YncE , which has similarity to the nonribosomal peptide synthase AfuA of Aspergillus fumigatus , is reduced more than 9-fold upon exposure of E . coli to the translational inhibitor puromycin [43] . We also determined that deletion of ybcJ results in a significant reduction in the incorporation of 35S-labeled methionine in vivo relative to wild type ( Figure 3G ) , indicating a decrease in the global rate of protein synthesis . Similarly , ribosome profile analysis ( Figure 3H ) showed that inactivation of yfgB decreased the level of mature polysomes actively engaged in mRNA translation and altered the cellular ratios of 30S and 50S ribosomal subunits relative to 70S monosomes . Moreover , both the ybcJ and yfgB mutants exhibited reduced translation fidelity ( Figure 3I ) as assayed by four reporter plasmids that measure the frequency of frameshifts and stop codon readthrough . Other orphans in this translation subnetwork include YibL , which copurified both with YfgB and YbcJ , and with RNA processing factors involved in ribosome biogenesis , such as the RNA pseudouridine synthetases RluB/RluC and the RNA helicase DeaD , and with RppH ( formerly NudH ) , which was recently identified as a regulator of 5′-end–dependent mRNA degradation [44–46] . Similarly , the orphan YdhQ copurified with translation elongation factor Tu , whereas YagJ interacted with lysine tRNA synthetase ( LysU ) ; and YjcF , which has similarity to phenylalanyl-tRNA synthetase PheT of Bacteroides vulgatus , bound ribosomal release factor 2 and another orphan , YbeB , which in turn was found to associate with the 50S ribosome subunit , as recently reported [47] . These results confirm that our high-confidence physical interaction network is informative about the function of at least certain orphans . Although we attempted to tag and purify the entire soluble E . coli interactome , we failed to detect 469 orphan proteins by MALDI or LCMS , presumably because they are membrane-associated ( ∼35%; Figure S1B ) and hence not soluble , or are of particularly low abundance ( ∼40% ) , as reflected by their CAI and mRNA levels ( Figures S1C and S1D , respectively ) . To bypass this limitation , we applied computational methods to discern a network of high-confidence pairwise functional interactions for all E . coli proteins , including those not detectable by proteomic methods , by examining the natural chromosomal clustering of bacterial genes . As illustrated in Figure 2B , we used four different GC methods , namely: ( 1 ) gene fusions [48 , 49]; ( 2 ) similarity between phylogenetic profiles [27 , 50 , 51]; ( 3 ) evolutionary conservation of gene order [52–54]; and ( 4 ) intergenic distances [55–57] ( see Materials and Methods and Protocol S5 for details ) . The latter two methods are independent approaches to detect operons and their subsequent rearrangements across prokaryotic genomes . In particular , the intergenic distances method , leads to considerably more high-quality predicted functional associations compared with the first three classic GC methods [55] , and does not depend critically on the detection of orthologs in evolutionarily distant genomes , making it potentially better suited for detecting functional interactions involving orphans . The pairwise interactions generated by each of these prediction methods were independently evaluated by benchmarking using suitable gold standards . Positive gold standards were defined as pairs of E . coli genes belonging to the same biological pathway as defined in EcoCyc , while the negative gold standards represented pairs of annotated E . coli genes whose products participate in different pathways ( see Protocol S5 for details ) . The results of each GC method were subsequently combined to create a single unified functional association score ( Protocol S6 and Figure 2B ) . Although different data integration algorithms have been developed [58–61] , most of these have a similar probabilistic basis and assumptions . For this study , we opted for the integration procedure used by von Mering and colleagues [61] to construct the Search Tool for the Retrieval of Interacting Genes/Proteins ( STRING ) database . This approach treated the reliability of the associations generated by each GC method as independent probabilities , such that the likelihood of an interaction is proportional to the number of times it was observed and the degree to which each GC method contributed to the overall network reliability ( Protocol S6 ) . Finally , we applied a stringent filter to the unified functional network to obtain a set of 74 , 776 high-confidence ( probabilities ≥0 . 80 ) nonredundant interactions ( Figure 4A and Table S8 ) . Despite the tendency of the orphans to exhibit more limited conservation notwithstanding the dependency of GC methods on homologs in multiple species ( except for operon predictions based on intergenic distances [55] ) , our combined GC network implicated virtually all ( 1 , 367 , or 96% ) of the orphans in 23 , 365 pairwise functional interactions ( Table S8 ) . Moreover , relatively few ( <18% ) of our predicted interactions appear to have been reported previously ( Figure 4B ) . Although we could not meaningfully compare our results to an alternate set of putative functional links generated recently [22] because of a lack of publicly accessible dataset scores , we found that less than 5% ( 3 , 368 ) of our predicted interactions are listed in the PROLINKS comparative genomics databank [62] , whereas only approximately 16% ( 11 , 842 , of which only 2 , 613 involve an orphan ) were found in STRING ( v . 7 . 1 ) at a more liberal 0 . 7 confidence threshold . More critically , greater than 85% of our predicted orphan interactions involve a functionally annotated E . coli protein , indicating a good potential to make functional inferences . The fact that PROLINKS has 1 , 657 predictions not attained by our integrative approach may reflect our use of a higher confidence threshold as well as differences in implementation of the GC measures and the identification of putative orthologs . For instance , whereas we used BLAST-BDBHs as criteria to detect orthologs between pairs of genomes , STRING uses COG-based definitions of orthology , whereas PROLINKS uses one-way BLAST hits ( not necessarily orthologs ) . Conversely , most of the 16 , 585 predictions exclusive to the STRING database were compiled using text mining or alternate experimental criteria such as protein–protein interactions , whereas the highest numbers of predictions exclusive to our GC datasets come from operon rearrangements . The reliability of our unified functional association network was independently corroborated based on the high correlations of expression among putatively interacting gene pairs ( Figure 4C ) , which was comparable to that observed for components of the same curated EcoCyc pathway even after eliminating all pairs of genes belonging to an experimentally characterized operon or all adjacent gene pairs in E . coli ( Figure 4C ) . We also observed a marked enrichment for interactions among proteins annotated to the same curated functional categories ( Figure 4D ) , implicating by extension any associated orphans in these same processes . Groups of functionally interacting genes form functional modules centered on a common process or biochemical pathway ( s ) . To define orphan participation as components of such modules , we partitioned the high-confidence GC network using MCL ( Protocol S4 ) , generating a total of 507 putative functional modules consisting of two or more components ( Figure 4E and Table S9 ) . Examination of the functional homogeneity of these predicted modules ( see Materials and Methods and Protocol S4 ) indicated , as for our putative multiprotein complexes , that they were highly enriched ( p <0 . 0001 ) for concerted annotated biological processes ( Figure S2D ) , again implicating the associated orphans in these same roles . Module membership followed a characteristic power law distribution ( Figure S2E ) with most modules having between two and 10 components , but the overall node connectivity did not ( Figure S2F ) ; further analysis is necessary to determine the significance of this divergent behavior . Two hundred and eighty-nine ( 57% ) of the modules had at least one of a total of 1 , 189 different orphans . One notable example is shown in Figure 4F . Diverse lines of experimental and bioinformatic evidence support the involvement of this putative module in the biogenesis and/or activity of fimbriae , appendages or pili that are shorter than the characteristic flagellum of Gram-negative bacteria , which mediate cell adhesion , biofilm formation , motility , and host invasion [63 , 64] . For instance , 12 of the 13 orphan components possess sequence characteristics of bacterial adhesins and chaperone/Usher pili protein families [29 , 65] . Gene expression profiling studies [66 , 67] have previously established that most of these orphans are also coordinately induced during biofilm formation ( Table S10 ) . Perhaps most compellingly , we found that single-gene E . coli knockout mutants of six of the 13 orphans display markedly reduced swarming capabilities in semisolid agar ( Figure 4G ) , while 11 out of 13 mutants were significantly impaired for biofilm formation in vitro as compared with a wild-type control ( Figure 4H ) . Taken together , these observations strongly implicate this set of orphans in the formation and/or proper function of fimbriae . Several other prominent modules are shown in Figure 4I . These comprise the orphans YdiN , YdiL , and YdiM predicted ( based on operon rearrangements ) to functionally interact with several members of the Aro operon known to participate in the metabolism of shikimate , a precursor of aromatic amino acids . Consistent with this , ydiN , aroD , and ydiB are reportedly overexpressed when E . coli is grown in media containing shikimate as the sole carbon source [68] . Moreover , we found that deletion of either ydiN or ydiB resulted in defective metabolism of shikimate causing phenotypic auxotrophy for aromatic amino acids ( Figures 4J and S3 ) as is observed for mutants of known aromatic amino acid biosynthetic genes ( e . g . , aroA and aroD ) [69] . Other functional modules include frlA/frlB , part of the Frl operon of E . coli responsible for the import and metabolism of the alternative carbon source fructoselysine , together with the orphan YifK , which has sequence characteristics of a transporter [38] , implicating it in electrochemical potential-driven uptake of this sugar . Conversely , two orphans , YecC and YecS , had functional associations consistent with linkages to amino acid and nucleotide metabolism , four ( YagU , YqeG , YhaO , and YhaM ) were linked to a putative module involved in transport and metabolism of threonine and serine , whereas three others ( YjjI , YeiM , and YjjJ ) were found in a module enriched for factors involved in nucleotide transport and degradation of deoxyribonucleosides . Taken as a whole , these results suggest discrete functional relationships for many previously unannotated proteins , implicating certain orphans within specific pathways . Examination of the extent of overlap between our physical and functional networks , both in terms of common binary interactions and shared components among the derived complexes ( from PI ) and modules ( from GC ) , indicated that they are largely complementary ( Table S11 ) . Since a similar trend was also evident comparing other existing curated E . coli PI datasets ( derived from either low-throughput or other high-throughput studies ) with independent GC inferences ( e . g . , from STRING; Table S11 ) , this presumably stems in part from the incomplete coverage obtained by these different approaches . Regardless , these observations imply that the union of PI and GC networks is necessary to capture the widest spectrum of biologically relevant interactions . Indeed , it has been shown previously that the combination of PI with functional genomic inferences , each statistically weighted according to dataset quality , can markedly improve both functional coverage and accuracy [59 , 61 , 70–72] . We therefore merged our experimental and predicted associations with the same method used to generate the unified GC network ( Figure 2C; see Materials and Methods and Protocol S6 for details ) . The resulting combined probabilistic network consisted of 80 , 370 high-confidence ( probability ≥75% ) putative pairwise interactions encompassing virtually the entire proteome of E . coli , including 2 , 769 ( 99% ) annotated proteins and 1 , 375 ( 96% ) functional orphans ( Table S12 ) . Graph analysis of this final integrated network ( Protocol S7 and Table S13 ) indicated that the orphans tended to have a lower overall connectivity and betweenness centrality , measured as the number of shortest paths going through a given node , relative to annotated components , suggesting more peripheral positions in the integrated networks . However , the orphans also exhibited lower average closeness , defined as the average length of shortest paths between any two nodes , and had similar overall clustering coefficients , indicating that , in general , the orphans are functionally connected to , rather than isolated from , the annotated gene products . These observations implied that consideration of both the individual associations and overall placement of the orphans within the integrated interaction network would facilitate functional deduction . We therefore devised a new network-based function prediction method ( termed StepPLR; see Figure 2C and Materials and Methods ) to exploit the global topological similarity among all the protein pairs and their corresponding functional annotations in the integrated network . Our method assigns functions to orphans based on the functional information from their first-order ( direct ) and second-order ( indirect ) annotated neighbors in the integrated network using penalized logistic regression models and a stepwise variable selection procedure to deduce optimal functional profiles ( see Protocols S8 and S9 for details ) . We based our classifications on the discrete COG functional categories and on the hierarchical , multifunctional terms of the Gene Ontology ( GO ) [73 , 74] and MultiFun classification schemas [25] . To avoid potential sources of false predictions , we removed any proteins labeled with the evidence codes IPI ( for inferred from protein interaction ) and IGC ( for inferred from genomic context method ) when generating the GO reference set , as well as proteins in poorly characterized categories in COGs and MultiFun ( Protocol S9 and Table S14 ) . As shown in Figure S4 , StepPLR had better precision and recall prediction performance than several other widely used guilt-by-association procedures tested , such as majority-counting and chi-square–based methods ( see Protocol S10 for details ) . Although the performance achieved for the different functional categories varied , our approach generated area-under-the-curve ( AUC ) values of 0 . 8 or higher for most of the COG ( 83% ) , GO ( 67% ) , and MultiFun ( 53% ) categories ( Table S15 ) , and was relatively insensitive to the number of annotated proteins per function . Moreover , since our method exploited the correlation among the different categories , most orphans had multiple biologically consistent predicted functions ( Table S16 ) . As displayed graphically in Figure 5A , our prediction procedure ultimately linked many of the orphans to specific , functionally related protein “neighborhoods” . We again made use of the MCL algorithm to objectively delimit functionally highly homogeneous ( p < 0 . 0001 ) protein groupings based on the profile similarity of annotations and predictions ( see Protocol S4 for details and Table S17 for listing ) . One notable example is the protein translation machinery ( Figure 5B ) , which has 23 associated orphans . To independently verify the functional relevance of these assignments , we examined the effects of deleting the corresponding genes in terms of conferring sensitivity to drugs that inhibit protein synthesis . Consistent with expectation of a direct role in protein synthesis , and similar to loss of bona fide annotated translation factors and tRNA synthetases , the mutant strains exhibited statistically significant ( p < 0 . 05 ) differential sensitivity as compared to wild type and unrelated gene mutants to a variety of antibiotics that selectively block protein translation ( Figure 5C and Table S18 ) . We also examined an alternate group of orphans ( YafP , YiaD , and YbcM ) associated with the flagellar biogenesis and motility apparatus ( Figure 5D ) . Single-gene knockout mutants of annotated components in this neighborhood exhibit decreased motility in semisolid agar as compared to wild-type E . coli strains [13] . Consistent with our functional predictions , we likewise found that deletion of yafP ablated cell motility in vitro ( Figure 5E ) , similar to mutants lacking core flagellum motor encoding genes ( e . g . fliH and fliM ) , whereas loss of yiaD and ybcM reduced swarming ( i . e . , decreased halo formation ) to an extent comparable to perturbation of other established flagellar components ( e . g . , flgJ and fliR ) . Additionally , a previous study [75] using phenotypic complementation analysis had suggested that a ybcM ortholog in Yersinia enterocolitica is likely an AraC-type regulatory protein involved in controlling bacterial motility . Taken together , these results suggest that , akin to several other recently discovered novel motility components [13 , 76] , these orphans are required for the proper assembly and/or subsequent locomotion of the E . coli flagella , a fundamental bacterial structure . Many other orphans were predicted to have roles in other conserved biological systems , such as DNA replication . For example , as shown in Figure 5F , we identified the orphan YhcG in association with DNA processing enzymes , including the restriction complexes HsdMRS and McrABC , the integrases IntF and IntS , and the recombinase PinE . YhcG has sequence characteristics of the PD- ( D/E ) XK superfamily of nucleases involved in DNA recombination and repair [77] . Consistent with these observations , we found that deletion of yhcG results in a synthetic-lethal phenotype ( Figure 5G ) when combined with hypomorphic alleles of the replicative primosome ( dnaB ) , DNA polymerase III ( dnaN ) , and DNA topoisomerase IV ( parE ) , consistent with a direct role in DNA replication or the resolution of critical intermediates . Our functional predictions were particularly revealing about bacterial cell envelope biology , with implications for infectious disease and antibiotic susceptibility . Like other free-living microbes , E . coli is encased in a membranous cell envelope composed of proteins , lipids , and carbohydrates that serves as the interface to its environment and mammalian host , yet over a third of the approximately 1 , 000 predicted membrane-associated and periplasmic proteins of E . coli are presently functionally unannotated [38] . Figure 6A shows a set of eight orphans linked to a functional neighborhood along with 29 annotated proteins with established roles in the biogenesis of peptidoglycan , a major structural component of the bacterial cell wall that is a target of many antimicrobials . Consistent with our functional predictions , E . coli cells deleted for these same orphans exhibited differential sensitivity to various antibiotics that inhibit peptidoglycan assembly ( Figure 6B ) . Moreover , the observed phenotypes were also characteristic of enzymes acting in the initial cytoplasmic and inner membrane stages of peptidoglycan biosynthesis , like murA and mtgA , rather than later periplasmic steps , like pbpC . This suggests orphans involvement in an early step in cell wall formation . Figure 6C shows a second functionally homogeneous neighborhood composed of 14 orphans and 91 annotated proteins associated with pathways linked to the biosynthesis of lipopolysaccharide ( LPS ) and other core cell envelope components . Consistent with these assignments , subcellular protein localization studies based on green-fluorescent reporter fusions [78] have previously established that at least five of these orphans ( YfbH , YfbJ , YfbW , YbjT , and YjdD ) are physically tethered to the E . coli inner membrane , whereas signal peptides potentially mediating export across the inner membrane have been predicted for YafL and YadE [79] . Further , the interactions of annotated and orphan components within this neighborhood suggest specific relationships consistent with particular biological pathways . For instance , YfbJ , YfbH , and YfbW interact with Ugd and four other annotated proteins that participate in maturation of the lipid A anchor of LPS . Previous work [80] had suggested that an unknown transporter ( s ) shuttles lipid A intermediates between the cytoplasm and periplasm . Coincidentally , YfbJ and YfbW are paralogs that belong to a superfamily of multidrug efflux transporters [29] , in which the C-terminus of YfbJ is cytoplasmic and that of YfbW periplasmic [78] . Since knockouts of the corresponding genes were recently reported to impair trafficking of lipid A precursors [81] , YfbJ and YfbW appear to be the relevant transporters . Likewise , YadE and three other orphans ( YjaH , YcaR , and YfhL ) are predicted to function in Rfa- , Lpx- , and Kds-based pathways participating in KDO2-lipid A ( i . e . , core LPS ) biosynthesis . Consistent with these predictions , we found that deletion of either yfbJ or yadE resulted in aggravating synthetic genetic interactions when combined with mutant alleles of genes in parallel pathways ( Figures 6C and 6D ) . Other orphans in this neighborhood ( e . g . , YiiD , YbjT , and YbcH ) are predicted to interact either with the Rff and Rfb pathways involved in the generation of other important cell envelope constituents , such as enterobacterial common antigen and O-antigen . Likewise , YibD and YafL interact with Wca and Wz proteins that participate in biosynthesis of colanic acid ( M-antigen ) . Moreover , although they are not formally classified within this route , we detected functional interactions of RfaD with LpxD , KdsA , and KdsC involved in the synthesis of ADP-l-glycero-d-manno-heptose , which is used by many enzymes for LPS production . Similar to mutants deficient for enzymes annotated to participate in the cell envelope biogenesis , we found that deletions of the orphans in this neighborhood significantly ( p < 0 . 05; see Protocol S11 ) perturbed cell viability upon exposure to antibiotics that block formation of the bacterial cell envelope ( Figure 6E ) , substantiating our functional predictions . To investigate the evolutionary significance of the putative functional associations detected in E . coli , we examined the presence of orthologs of each of the interacting orphan and annotated protein pairs among currently available prokaryotic genomes ( see Materials and Methods and Protocol S12 for details ) . As might be expected from the more limited evolutionary scope of the orphans ( cf . Figure 1G ) , functional interactions involving orphans were typically less broadly distributed than those of annotated proteins ( Figure 7A ) , with the least-frequent distribution category consisting of interactions between the orphans themselves . Nevertheless , our analysis indicated extensive conservation of orphan associations across all sequenced forms of prokaryotic taxa . For instance , 5 , 553 putative interactions between orphan and annotated proteins , and 603 among orphan pairs alone , were predicted to be distributed as far as Archaea , again supporting the importance of orphans beyond that anticipated from the biases evident in previous functional characterizations of E . coli . Moreover , although it might be expected that GC predictions involving gene-order conservation and phylogenetic profile similarity would be biased towards highly conserved genes , both the PI ( independent of gene conservation ) and functional interactions ( with only operon rearrangement–based predictions less dependent on gene conservation ) follow the same tendencies . Hence , if there was residual bias towards finding functional interactions for only the most highly conserved genes , it is not pronounced . We next compared the phylogenetic distributions of the orphan and annotated components of the 97 predicted functional neighborhoods with at least one orphan ( Figure 7B ) . Our results indicate that only four neighborhoods have orphans that are more widely distributed than annotated members . An example , consisting of three highly conserved orphans linked to metabolism and one poorly conserved annotated component is shown in Figure 7C . This pattern suggests the orphans are of primary functional significance . In contrast , most ( 57 ) other neighborhoods , including several described in previous sections ( cf . Figures 5B , 5D , 6A , and 6C ) , generally have similar distribution profiles of orphan and annotated members ( Pearson correlation >0 . 5 ) , suggesting equal participation across diverse biological processes despite the incomplete previous classifications . For example , the average component distributions of one such representative neighborhood involved in drug efflux is shown in Figure 7D . Six of the seven orphans of this particular neighborhood are predicted to localize to the E . coli inner membrane [38] . We found that deletion of these orphans results in hypersensitivity to an otherwise exported drug , similar to that observed upon loss of the annotated components ( Figure 7E ) , suggesting equal participation in the maintenance of cell homeostasis . Conversely , annotated components exhibited a broader phylogenetic distribution in the remaining 36 neighborhoods . For example , two orphans ( YneG and YdaU ) linked to DNA replication ( Figure 7F ) were far more evolutionarily restricted than their annotated counterparts . Nevertheless , deletion of these same genes markedly reduced cell viability upon exposure to an inhibitor of DNA replication ( Figure 7G ) similar to hypomorphic strains of their broadly distributed replication partners DnaT , DnaA , and DnaE . Therefore , the sparser distribution of these orphans , and those of the other 35 neighborhoods , may reflect roles as critical accessory factors in their respective biological systems , perhaps to fine tune cellular responses to particular environmental adaptations and selective pressures ( e . g . , exposure to antibiotics ) .
Defining the precise biological roles and relationships of bacterial gene products in an often dynamically changing physiological context is a challenging proposition . Historically , systematic assessments of protein function in bacteria have tended to rely on molecular inferences based on sequence alignments and domain architectures , whereas experimental characterization has traditionally been driven by specific scientific interests rather than with the aim of providing the broader community with unbiased collections of functionally related proteins and phenotypes . Since the biological role of a protein is not necessarily reflected in its primary sequence , the elucidation of molecular interaction networks can provide an alternate perspective even in the absence of detailed phenotypic data [16 , 71] . Here , we have opted to view a model microbial cell mechanistically as a series of modular molecular interaction networks that underlie the major biochemical processes that mediate cell homeostasis and proliferation , wherein the functional attributes of particular gene products are reflected in their overall patterns of associations . To this end , we have generated an extensive compendium of physical and functional linkages covering almost the entire protein-coding complement of E . coli . This led to the elucidation of hundreds of putative soluble multiprotein complexes and functional modules encompassing virtually all the many gene products currently lacking public annotations . Although existing integrative probabilistic interaction databases such as STRING [61] and EcID [82] provide valuable additional binary interactions that are potentially useful for protein function prediction or as complementary evidence to those reported in this study , our machine learning strategy goes beyond describing binary interactions by explicitly describing the most probable biological functions of the orphans . Of particular noteworthiness , our functional predictions and phylogenetic projections associate a sizeable fraction of the functional orphans with core bacterial processes , suggesting they may have previously eluded detection in part due to prior analytical biases . Since the various methods used in this study to discover different types of molecular relationships also have their own intrinsic biases , complementary information was obtained through data integration . The limited overlap between the high-confidence physical and functional interaction networks presumably stems in part from the incomplete coverage typically achieved by high-throughput experiments and their methodological differences [13 , 83] . For example , certain orphans were difficult to evaluate by GC methods due to a lack of apparent orthologs at medium-to-high evolutionary distances , which hinders comparative genomic inferences . Likewise , although we performed large-scale tandem affinity tagging and purification under near-native physiological conditions to generate highly purified preparations of stable , endogenous multiprotein complexes , we did not achieve complete coverage of the proteome . We did not attempt to purify a large number of membrane-associated proteins , which require specialized solubilization procedures , whereas the soluble proteins that we failed to tag or detect by mass spectrometry were presumably either of very low abundance or not expressed in our growth conditions . Comparison of our physical interaction network with analogous public datasets produced for other model species , such as worm , fly , yeast , and even the bacterium Helicobacter pylori , revealed very limited ( <1% ) overlap . These observations are congruent with recent findings by Rajagopala and colleagues [13] showing that only a third ( 49 ) of the 173 experimentally derived PI in the cell motility network of the spirochete Treponema pallidum predicted to occur in the ε-proteobacteria Campylobacter jejuni on the basis of orthology could subsequently be confirmed by targeted two-hybrid testing . The limited overlap between proteomic datasets presumably reflects a combination of incomplete coverage by various experimental assays , methodological differences and evolutionary divergence . The observation that the intersection of functional genomics inferences with low-throughput curated physical interaction data is somewhat higher ( cf . Table S11 ) might be explained by two nonmutually exclusive ways: first , protein–protein interactions reported in the literature based on traditional biochemical methods might be biased towards the most evolutionarily conserved multiprotein complexes , which tend to be enriched for essential components with broadly distributed phylogenetic profiles that are more easily and accurately predicted by GC methods; second , the relatively high sensitivity of the two complementary forms of protein mass spectrometry used in this study may have resulted in the detection of lower-abundance orphan proteins that have previously not been studied in depth . The last point is consistent with the notion that different proteomic methods capture different PI types [83] . Hence , alternative proteomic methods , such as two-hybrid screens [13 , 84–86] or in vivo protein-fragment complementation assays [87] , may be better suited for detecting certain PI currently underrepresented in our dataset . In a similar vein , additional functional relationships will undoubtedly be uncovered by different experimental and computational procedures , such as high-throughput comparative analysis of mutant cellular phenotypes [2] , genome-wide genetic interaction screens [88 , 89] , and automated text mining [90 , 91] . The topological properties inherent to biological networks ( e . g . , their hierarchical organization and degree distributions ) combined with incomplete interactome coverage make establishing definitive functional groupings difficult [92] . Our approach was to take into account both the correlations among functional categories and the overall topological structure of the integrated network to generate a more balanced probabilistic model . Whereas alternative methods may provide enhanced interpretations of the organizational properties of the PI and GC networks , the functional enrichment and experimental validations established here suggest that our network-based computational inferences provide a reasonable perspective for exploring bacterial protein function . Similar strategies have resulted in powerful predictors of protein function in Eukaryotes [49 , 72 , 93–95] . The potential tradeoff is that additional error or uncertainty may have occasionally been introduced by assuming functional similarity among more loosely connected proteins . Moreover , the probabilities associated with particular functional terms may not be directly comparable . Functional orphans associated with very well-characterized biological processes are more likely to be correctly assigned by computational methods [72] , whereas those associated with relatively poorly studied pathways will tend to remain obscure . Nonetheless , they can be grouped together on the basis of specific PI , GC , or even other functional associations ( cf . Figure 7C ) and hence serve as functional groupings rather than isolated entities . In general , the high-confidence functional relationships we inferred for E . coli could be validated by independent experimental tests , and can be extrapolated to other bacterial species , including pathogens . Over 35% of the orphans find orthologs as far away as Archaea , and hence are likely associated with the same basic housekeeping processes we predict for E . coli , such as formation of the cell wall and protein synthesis . For instance , we have established putative roles in sugar and lipid metabolic pathways for several dozen evolutionarily conserved orphans that appear to be critical for proper biogenesis of the bacterial cell envelope , and hence may represent novel targets for antibiotic development . Conversely , our systematic comparisons also revealed some unique aspects of the orphans in the evolutionary history of E . coli , such as the potential fimbriae factors that appear to be restricted to Enterobacteriaceae . One interpretation is that orphans with limited phylogenetic distributions contribute to fine tuning of adaptive physiological responses upon changing environmental conditions , as previously suggested for peripheral metabolic genes acquired by horizontal transfer [96] . Conversely , the fact that the interactions of orphans with annotated proteins show a higher proportion of conservation across taxa implies that conserved biological systems are still to be discovered , and whose member contributions could extend across evolutionary domains . The physical and functional associations reported here are therefore presented as a Web-accessible public resource called “eNet” ( http://ecoli . med . utoronto . ca; see Protocol S13 for details ) to facilitate exploration of the fundamental molecular biology of bacteria in general and for hypothesis-driven studies of unique aspects pertaining to E . coli more specifically .
Large-scale SPA tagging and purifications were performed essentially as previously described [4 , 32] . Briefly , a DNA cassette encoding the SPA-tag and a selectable marker flanked by gene-specific targeting sequences was amplified by PCR using primers with homology to a selected locus . The cassette was then transformed and integrated using homologous recombination in the lysogenic E . coli strain DY330 ( W3110 background ) , which harbors the highly efficient λ-phage–encoded homologous recombination enzymes exo , bet , and gam under the control of the temperature-sensitive CI857 repressor ( the “Red” system ) , to create a C-terminal fusion with the protein of interest . Strains in which the PCR product was integrated were subjected to antibiotic selection , and tagged protein expression was confirmed by western blotting . Tagging primer sequences are available upon request . Two complementary mass spectrometry techniques ( gel-based MALDI peptide mass fingerprinting and gel-free LCMS shotgun sequencing ) were used to detect physically interacting proteins . Details about the large-scale strain culture , protein extraction and purification , and protein identification procedures are provided in Protocol S3 . Scoring of tentative PI from the LCMS and MALDI assays was conducted using a logistic regression model using reference PI obtained by low-throughput experiments curated in the DIP , BIND , and IntAct databases [35–37] as a positive training set . Our negative training set consisted of pairs of proteins in which one component was experimentally determined or predicted with high confidence to be cytoplasmic and the other residing in the outer membrane or the periplasm [38]; inner membrane proteins were discarded from this negative dataset since they are in physical proximity ( and hence could potentially physically interact ) to cytoplasmic and periplasmic proteins . Our logistic regression procedure also took into account the degree of consistency of copurifying protein pairs , balancing the tradeoff between “spoke” and “matrix” representation models of interactions within copurified groups of proteins to decrease the false discovery rate . We then combined the scores derived from LCMS and MALDI into a a single PI network using a previously established procedure for integrating probabilistic networks [61] , which assumes the reliabilities of associations generated by these methods are independent ( see Protocol S6 for details ) . To facilitate independent critical evaluation , all our processed interaction data is available through the Web site in HUPO-PSI molecular interaction reporting format ( standard level 2 . 5 ) . The four GC methods used to predict functional interactions among E . coli proteins were based on: ( 1 ) functional linkages among genes which fuse to form a single open reading frame in at least one other genome , i . e . , gene fusion [48]; ( 2 ) the mutual information of the coordinated presence or absence of pairs of genes across a set of 440 nonredundant genomes , i . e . , phylogenetic profiles [51 , 97]; and the natural chromosomal association of bacterial genes in operons as detected by two alternative methods , namely ( 3 ) the tendency of genes forming operons to show small intergenic distances [98 , 99] , and ( 4 ) the conservation of gene order , in which a confidence value for each pair of adjacent genes in the same strand was used as indicator that those genes likely form an operon , as compared with the conservation of adjacent genes in opposite strands [53] . For the last two methods , subsequent operon rearrangements were detected by genomic mapping of orthologs across 440 nonredundant bacterial genomes [55] . For all four GC methods , we used the BLAST-BDBHs as an operational definition of orthology ( see Protocol S5 for details ) . To avoid circularity , the prediction scores of the four GC methods were benchmarked separately using proteins belonging to the same metabolic pathway according to EcoCyc [11] as positive reference set , and proteins in different pathways as negatives ( Protocol S5 ) . A single , unified high-confidence functional association network was then constructed by integrating the interaction predictions generated by the four GC methods using the same scoring model [61] used to integrate the MALDI and LCMS data ( Protocol S6 ) . Protein clusters were generated from three different networks using MCL [40] ( Figure 2 ) : ( 1 ) the PI network ( generating protein complexes ) ; ( 2 ) the unified GC network ( generating functional modules ) ; and ( 3 ) the function prediction/annotation profiles derived from the integration of PI and GC networks ( generating functional neighborhoods ) . The core idea of MCL is to simulate random walks ( i . e . , flux ) among the proteins ( nodes ) within each network to delimit regions with high flux , taking into account the connectivity and weight of interaction edges . In this work , edge weights correspond to the likelihood of pairwise protein interactions in each network . In each case , the global MCL inflation parameter , which tunes the granularity of the delimited clusters , was optimized by balancing the mass fraction of clusters and efficiency of partitions ( see Protocol S4 for details ) . The resulting clusters were individually assessed for functional homogeneity in terms of COG annotations as described previously [100] ( Protocol S4 ) . Our algorithm ( StepPLR ) for assigning biological functions is essentially a network topology–based method in which the functions of the orphans are predicted based on the functions of their associated annotated proteins in the immediate ( direct ) and adjacent ( indirect ) network vicinity ( see Protocol S9 for details ) . Briefly , a single network integrating the high-confidence PI and GC probabilistic networks was first created using the same scoring model [61] used to integrate the PI data and the four GC networks . The weighted topological overlap [101] between each pair of protein nodes in the integrated network was then calculated to determine the correlated functional profiles based on a penalized logistic regression model ( see Protocol S8 for details ) . Finally , a stepwise variable selection procedure to optimize function profiles in the final logistic regression was used ( see Protocol S9 for details ) . Only functional categories with at least 15 annotated E . coli proteins were used in our integrated functional association network ( see Table S14 ) : 18 COG classes , corresponding to bacterial protein functions; 19 biological classes from MultiFun , in which the proteins can have multiple annotations based on different classification criteria; and 51 biological process classes in GO . Other guilt-by-association representative methods ( e . g . , majority-counting and chi-square–based ) were also evaluated ( results shown in Figure S4A ) . Expanded descriptions of benchmarking and other computational procedures of our function prediction algorithm are provided in Protocols S9 and S10 . Orphans were selected for experimental validation of functional predictions based on the following criteria: ( 1 ) the orphan was predicted to perform a function for which a suitable phenotypic assay was previously reported ( e . g . , an antibiotic targeting the associated function was available ) ; ( 2 ) the orphan was clearly grouped with select annotated genes , allowing the inclusion of positive as well as negative controls; and ( 3 ) the orphan had high ( >0 . 8 confidence ) function prediction score ( s ) . Antibiotic susceptibility assays were performed by pinning orphan and annotated gene knockout mutants [2] onto solid media plates in the presence or absence of antibiotics , and then imaging and comparing colony sizes . Details of the antibiotic sensitivity , translation , and auxotrophy assays are provided in Protocol S11 . Motility assays were performed with overnight E . coli strain cultures pinned onto rectangular Petri dishes ( Singer ) containing semisolid swarming agar ( LB medium with 0 . 25% agar ) . The swarming phenotype was classified visually based on the cell spreading-halo diameter observed after approximately 8 h incubation at 32 °C . Biofilm formation assays were conducted essentially as described in [102] , with replicate data normalized relative to wild-type controls ( Protocol S11 ) . Epistatic genetic interactions between pairs of gene mutants in E . coli were identified using a newly developed conjugation-based screening method [88] . Briefly , a drug resistance–marked query gene deletion in a high-frequency recombination donor strain was crossed into either single-gene deletion knockout mutants from the Keio strain collection [2] or select essential gene hypomorphs to generate double mutants . After double drug selection , synthetic lethal or sick phenotypes were scored visually according to measured colony sizes ( Protocol S11 ) . | One goal of modern biology is to chart groups of proteins that act together to perform biological processes via direct and indirect interactions . Such groupings are sometimes called functional modules . The types of protein interactions within modules include physical interactions that generate protein complexes and biochemical associations that make up metabolic pathways . We have combined proteomic and bioinformatic tools , and used them to decipher a large number of protein interactions , complexes , and functional modules with high confidence . In addition , exploring the topology of the resulting interaction networks , we successfully predicted specific biological roles for a number of proteins with previously unknown functions , and identified some potential drug targets . Although our work is focused on E . coli , our phylogenetic projections suggest that a considerable fraction of our observations and predictions can be extrapolated to many other bacterial taxa . As all the data derived from this study are publicly available , others may build on our work for further hypothesis-driven studies of gene function discovery . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biotechnology",
"biochemistry",
"computational",
"biology",
"evolutionary",
"biology",
"genetics",
"and",
"genomics"
] | 2009 | Global Functional Atlas of Escherichia coli Encompassing Previously Uncharacterized Proteins |
The progressive decline of CD4+ T cells is a hallmark of disease progression in human immunodeficiency virus ( HIV ) and simian immunodeficiency virus ( SIV ) infection . Whereas the acute phase of the infection is dominated by virus-mediated depletion of memory CD4+ T cells , chronic infection is often associated with a progressive decline of total CD4+ T cells , including the naïve subset . The mechanism of this second phase of CD4+ T cell loss is unclear and may include immune activation–induced cell death , immune-mediated destruction , and regenerative or homeostatic failure . We studied patterns of CD4+ T cell subset depletion in blood and tissues in a group of 20 rhesus macaques inoculated with derivatives of the pathogenic SIVsmE543-3 or SIVmac239 . Phenotypic analysis of CD4+ T cells demonstrated two patterns of CD4+ T cell depletion , primarily affecting either naïve or memory CD4+ T cells . Progressive decline of total CD4+ T cells was observed only in macaques with naïve CD4+ T cell depletion ( ND ) , though the depletion of memory CD4+ T cells was profound in macaques with memory CD4+ T cell depletion ( MD ) . ND macaques exhibited lower viral load and higher SIV-specific antibody responses and greater B cell activation than MD macaques . Depletion of naïve CD4+ T cells was associated with plasma antibodies autoreactive with CD4+ T cells , increasing numbers of IgG-coated CD4+ T cells , and increased incidence of autoreactive antibodies to platelets ( GPIIIa ) , dsDNA , and phospholipid ( aPL ) . Consistent with a biological role of these antibodies , these latter antibodies were accompanied by clinical features associated with autoimmune disorders , thrombocytopenia , and catastrophic thrombotic events . More importantly for AIDS pathogenesis , the level of autoreactive antibodies significantly correlated with the extent of naïve CD4+ T cell depletion . These results suggest an important role of autoreactive antibodies in the CD4+ T cell decline observed during progression to AIDS .
Progressive CD4+ T cell decline in human immunodeficiency virus ( HIV ) infection , the slow but persistent loss of both naïve and memory CD4+ T cells [1]–[3] , is an important marker of progression to AIDS . The mechanism of chronic CD4+ T cell depletion is not clear since the number of infected cells at any one point in time would not seem to account for the extent of CD4+ T cell loss . Such progressive decline in CD4+ T cells is also observed in some but not all simian immunodeficiency virus ( SIV ) macaque models for AIDS [4]–[6] . Recent studies have shown that selective memory CD4+ T cell depletion is characteristically observed during the acute stage of SIV- and HIV-infection , presumably through direct killing of CCR5+CD4+ target cells [7]–[12] . However , early mucosal depletion of memory CD4+ T cells is also observed in nonpathogenic infection of natural African monkeys such as African green monkeys and sooty mangabeys [13] , [14] . Additional events are therefore required for AIDS pathogenesis in the pathogenic SIV/macaque model and in HIV-infected humans . Many host-mediated mechanisms have been suggested for the depletion of CD4+ T cells , including failure of regeneration and homeostasis , immune activation-induced cell death , autoimmune destruction and disruption of the lymphoid microarchitecture by collagen deposition [15]–[18] . Although HIV-1 is named for the immunodeficiency it induces , chronic immune activation is also a characteristic feature of this disease [1] , [6] , [19]–[22] . The chronic stages of HIV-infection , as well as pathogenic SIV infection , are characterized by generalized lymphadenopathy , immune activation of T-lymphocytes , hyper-gammaglobulinemia , and polyclonal B cell hyperactivity . SIV infection of natural host species , such as sooty mangabeys and African green monkeys , is not associated with the abnormal chronic immune activation and the accelerated T cell turnover seen in pathogenic models , suggesting a critical role of immune activation in the pathogenesis of AIDS [23] , [24] . Chronic SIV-infection of macaques provides a relevant and useful model to explore the mechanisms for progressive CD4+ T cell loss in AIDS pathogenesis . Earlier studies in our lab demonstrated that the rate of disease progression was associated with distinct patterns of CD4+ T cell decline in SIV-infected macaques [6] . SIV-infected macaques that progress rapidly without adaptive immune responses show the most severe memory CD4+ T cell loss with relative preservation of naïve cells [6] , [25]–[30] . In contrast , a progressive loss of total CD4+ T cells including naïve cells is observed during chronic SIV infection of conventional progressors [4]–[6] , as is also observed in HIV-1 patients [1]–[3] . Since the level of CD4+ T cells in the blood of HIV-infected patients is predictive of the onset of AIDS , we wished to explore this second pattern in greater detail . In this study , we used a SIV/macaque model to explore the mechanisms of CD4+ T cell depletion , and found that the progressive loss of CD4+ T cells was associated with antibodies that reacted with CD4+ T cells . This pattern of depletion was observed in a subset of animals with intense immune activation and autoimmune manifestations consistent with a functional role of these antibodies in the observed CD4 depletion .
To clarify the mechanism of progressive CD4+ T cell decline in the SIV/ rhesus macaque model , we analyzed CD4+ T cell naïve and memory subsets in retrospective samples of blood and various tissues of twenty macaques infected with derivatives of the pathogenic SIVsmE543-3 or SIVmac239 ( Table 1 ) . We found that the pattern of depletion of naïve and memory CD4+ T cell subsets in terminal blood samples differed substantially , showing two patterns of depletion , predominantly affecting either naïve or memory CD4+ T cells ( Figure 1A and Figure S1 ) . Animals were classified at death as either ND ( naïve depleted ) or MD ( memory depleted ) based upon the ratio of naïve to memory CD4+ T cells in the blood ( Figure 1B ) using a ratio of less than three to identify ND macaques . This classification was confirmed by the naïve to memory ratio in the peripheral lymph nodes ( PLN ) and spleen of 12 animals ( Figure 1B ) . Twelve macaques showed primarily naïve cell depletion and eight macaques exhibited memory cell depletion ( Table 1 and Figure 1A ) . The naïve to memory ratio of these two groups of animals differed significantly from one another in peripheral blood mononuclear cells ( PBMC ) , PLN and spleen at death ( Figure 1B: P = 0 . 0002 , 0 . 0025 and 0 . 0006 ) , showing much higher ratios in MD macaques . Comparison of CD4+ T cell subsets at pre-inoculation and at death revealed a significant decline in the memory subset in MD macaques and in both subsets in ND macaques ( Figure 1C ) . Figure 2A shows flow cytometric data for a representative macaque of each group . Animal H723 with naïve cell depletion showed a fairly normal representation of subsets in PBMC but preferential depletion of naïve CD4+ T cells in lymphoid tissues , where naïve cells are normally abundant in healthy macaques ( Figure 2A and 2B ) . Conversely , H718 , a macaque with primarily memory CD4+ T cell depletion showed selective depletion of the CD4+ memory subset in all tissues ( Figure 2A and 2B ) . As shown in Figure 2C , both memory and naïve CD4+ T cell populations gradually declined in H723 during the course of infection , but only the memory population declined in H718 . Due to the predominance of naïve cells in peripheral blood in H718 , a progressive decline of total CD4+ T cells in the chronic phase was only observed in H723 . Similar patterns in the kinetics of naïve and memory CD4+ T cell loss were observed in the remaining animals of each group . These data were confirmed by examining the kinetics of CD4+ T cell declines in 12 individual animals ( n = 8 ND and n = 4 MD ) as shown in Figure 3 . A progressive loss of CD4+ T cells , a slow decline of both naïve and memory subsets , was observed in the ND macaques ( Figure 3; left panels ) . Although a precipitous decline in all CD4 subsets ( as well as CD8+ T cells and B cells , data not shown ) occurred terminally in many of the MD macaques possibly due to acute regenerative failure , MD macaques primarily exhibited selective memory CD4+ T cell depletion ( Figure 3; right panels ) . Both groups contained macaques infected with a variety of inocula with the exception of SIVsmH635FC-infected macaques that were all classified to the ND group ( Table 1 ) . Thus , it appears that the inoculum was not a major factor in determining the type of disease course . These two groups of macaques differed significantly in terms of viral replication and humoral immune responses . As shown in Figure 4A , plasma viral load at the time of death was significantly higher in MD than ND macaques . Conversely , anti-SIV specific antibody titers , as measured by ELISA , were significantly higher in ND than MD macaques ( Figure 4B ) . The higher viral load , low antibody responses and profound memory CD4+ T cell depletion seen in the MD macaques , are all characteristic of a syndrome of rapid progression ( RP ) macaques that is commonly observed in a subset of macaques infected with pathogenic strains of SIV [25]–[30] . Consistent with this similarity , MD macaques included five RP macaques and three macaques with a longer disease course ( Table 1 ) . In contrast , none of the ND macaques were typical rapid progressors as defined by lack of SIV-specific antibody responses . However , in at least seven of the ND animals , the disease course was interrupted prior to the development of opportunistic infections by adverse events such as thrombosis . SIV-specific cellular immune responses were not assessed in this cohort but are likely to provide another potential discriminator between these two groups since rapid progression is also associated with failure to maintain SIV-specific CTL responses [30] . The median survival of the MD group was significantly shorter than the ND group ( 29 . 5 and 70 . 5 weeks respectively , P = 0 . 0135 ) . In terms of pathologic outcome , opportunistic infections were observed in both ND and MD macaques . As expected from the association of SIV encephalitis and pneumonia with rapid progression , these findings were more common in the MD group [25] . A pathologic finding that was unique to the ND group was the occurrence of thrombosis of major vessels such as the pulmonary artery , and vena cava; these events resulted in the acute death of six of the animals ( H596 , H704 , H709 , H714 , H723 and XGE ) in this group . Secondary indicators of immune activation such as intestinal amyloidosis ( n = 2 ) , generalized lymphadenopathy ( n = 12 ) and lymphoproliferative syndrome ( n = 1 ) , were commonly observed in the ND group ( Table 1 ) . As a measure of lymphocyte proliferation in lymph nodes , immunohistochemistry for Ki-67 expression was performed on sequential lymph node biopsies . As shown in Figure 5A , this analysis demonstrated increasing proliferation in T cell areas and the germinal centers of ND macaques . The development of follicular hyperplasia was a particularly prominent feature of lymphoid tissue in ND macaques and this was often accompanied by increasing numbers of B cells in the blood ( Figure 5B ) . Although immune activation , as indicated by increased levels of Ki-67+ CD4+ T cells , was observed early in infection in all of the animals , the number of proliferating CD4+ T cells , were higher in ND than MD macaques during the chronic phase of infection ( Figure 6A ) . To examine whether the immune activation in ND macaques was induced by increased microbial translocation across the intestinal lumen [19] , we measured lipopolysaccharide ( LPS ) levels in terminal plasma samples ( Figure 6B ) . Both the ND and MD animals showed elevated levels of LPS ( mean; 18 . 6 and 8 . 8 pg/ml , respectively ) as compared to uninfected rhesus macaque plasma LPS ( mean; 1 . 14 pg/ml ) . An outlier with extremely elevated LPS resulted from gram-negative bacterial sepsis ( H679 ) . While there was no statistical significant difference , LPS levels in ND macaques did trend higher than in MD macaques . The level of soluble CD14 , which is secreted from CD14+ monocyte/macrophages in response to chronic LPS stimulation , was also elevated in both groups ( Figure 6C ) , consistent with significant microbial translocation in both groups . The lack of proliferation of CD4+ T cells in MD macaques terminally suggests that these animals have lost the ability to respond to LPS stimulation due to profound immunosuppression . As is also commonly observed in HIV-infection [31] , [32] , many SIV-infected macaques exhibited thrombocytopenia during their disease course [33] , [34] . As shown in Figure 7A , a progressive decline was observed throughout the course of infection in ND macaques and the platelet count at the time of death was significantly lower in ND than MD macaques ( Figure 7B ) . Antibodies against platelet glycoproteins were detected in ND macaques more frequently than in MD macaques ( Figure 7C ) . Similar to immune thrombocytopenia in HIV infection , anti-GPIIIa was the most frequently observed platelet autoantibody [31] . Many of the macaques with severe thrombocytopenia died suddenly due to thrombosis of a major vessel and this finding was unique to ND macaques . Although such thrombotic events and associated arteriopathy are commonly associated with chronic SIV-infection , their etiology is unclear . In HIV-infected patients , a hypercoagulable state induced by autoantibodies to phospholipid ( aPL ) and other clotting factors can lead to thrombotic events [35] . We therefore examined plasma samples from these macaques for the presence of aPL antibodies . As shown in Figure 7D , a significant increase in aPL was observed in ND but not MD macaques . ND macaques also exhibited elevated dsDNA antibodies , suggesting a generalized autoimmune dysregulation in ND macaques ( Figure 7E ) . Consistent with a functional consequence in macaques , aPL titers inversely correlated with platelet counts and aPL titers were significantly higher in macaques with catastrophic thrombotic events ( Figure 7F and 7G ) . The memory CD4+ T cell loss in MD macaques would appear to be mainly explained by direct cell killing of CCR5+CD4+ target cells that is compounded by insufficient production [7] , [8] , [29] . In contrast , the loss of naïve CD4+ T cells in ND macaques was not readily explained by virus killing since these cells primarily express CXCR4 , not CCR5 [11] and SIV strains generally use CCR5 as their major co-receptor . The emergence of a CXCR4-tropic SIV has only been observed once in macaques and was associated with extensive substitutions in the V3 region of Env [29] , [36] . Similarly , emergence of X4-tropic SIVs in sooty mangabeys was associated with changes within the V3 loop [37] . Our prior sequence analysis of sequential plasma virus from a number of these animals did not show evolution of this region of Env to X4 tropism [27] ( Figure S2 ) . Therefore we used envelope clones from 20 wpi plasma of two of these animals ( H704 and H709 ) to evaluate whether emergence of a X4 variant explained the naïve CD4+ T cell depletion observed in these animals by this time point . Virus pseudotypes produced by cotransfection of appropriate envelope clones with pSG3ΔEnv , a Rev expression plasmid , were assayed for sensitivity to the CCR5 and CXCR4 antagonists , TAK779 and AMD3100 in TZM-bl cells . As observed in Figure 8 , the parental viruses SIVsmH635 and SIVsmE543-3 and clones derived from two ND macaques were inhibited only by the CCR5 antagonist , consistent with maintenance of CCR5 as their major co-receptor . This contrasted with the expected inhibition of the CXCR4-tropic HIV-NL4-3 by AMD3100 . To validate this finding , we also evaluated the infection frequency by quantitative real time PCR for SIV DNA within sorted populations of naïve and memory CD4+ T cells from samples of spleen collected at necropsy of five of the ND macaques . By this time point , only one sample ( H709 ) had sufficient naïve cells ( 8 . 8% ) for accurate cell sorting and qPCR . For this sample , the ratio of SIV gag copies per cell number was 29% of memory CD4+ T cells versus 5% of naïve CD4+ T cells , consistent with preferential infection of memory CD4+ T cells . Hence the emergence of CXCR4-tropic SIV within ND macaques did not appear to explain the depletion of naïve CD4+ T cells in at least two of these animals . Further more detailed analysis would be necessary to totally eliminate a role for X4 viruses . The heightened immune activation and the propensity for autoimmune manifestations in the ND macaques led us to explore the possibility that autoimmune mechanisms might account for the loss of naive CD4+ T cells in these animals . We therefore examined terminal plasma samples for antibodies that bound the surface of CD4+ T cells from healthy , uninfected donor macaques . In addition , terminal CD4+ T cells of study macaques were analyzed for surface IgG and IgM . As shown in Figure 9A and 9B , antibodies that bound CD4+ T cells were detected exclusively in plasma or purifed IgG from ND macaques . Moreover , the level of anti-CD4+ T cell IgG correlated significantly with B cells count at 16 wpi , when activation of B cells were apparent ( Figure 9C and Figure 5B ) . Most importantly , the level of anti-CD4+ T cell IgG correlated significantly with a lower naïve/memory ratio of CD4+ T cells , an indicator of naïve depletion ( Figure 9D ) as well as with absolute naïve CD4+ T cell counts at death ( Figure 9E ) . We next examined T cells from PBMC samples from ND and MD macaques for the presence of surface-bound IgG by flow cytometry . In vivo antibody binding was also exclusively detected on the surface of CD4+ T cells from ND macaques , as shown by representative flow cytometry plots in Figure 10A . The percentage of CD4+ T cells with surface IgG was significantly higher in the ND versus the MD group ( Figure 10B ) . In vivo binding of IgM was also detected on the CD4+ T cells from ND macaques , though there was no significant difference between the two groups ( Figure 10B ) . Similar to plasma antibodies , a significant inverse correlation was observed between the percent of IgG+CD4+ T cells and the memory/naïve CD4+ T cell ratio ( Figure 10C ) ; macaques with higher anti-CD4+ T cell antibodies had lower naïve/memory ratio , indicative of naïve CD4+ T cell depletion . The kinetics of the development of CD4+ T cell bound IgG and IgM were analyzed on cryopreserved PBMC samples ( Figure 11 ) . IgG was detected on the surface of CD4+ T cells in most of the ND macaques by eight or 12 wpi , peaking generally around 30 wpi . The emergence of autoantibodies was coincident with the development of naïve CD4+ cell deletion . IgM was also detected on CD4+ T cells but at much lower levels , gradually increasing in most of ND macaques . Similar results were obtained from the analyses of CD8+ T cells; antibodies that bound CD8+ T cells from SIV-uninfected macaques were detected in plasma and on the surface of CD8+ T cells of ND macaques ( Figure 12A ) but the levels were lower than that observed on CD4+ T cells . Depletion of naïve CD8+ T cells , which is common in HIV-infected patients [1] , [2] , was specifically observed in ND macaques ( Figure 12B ) , suggesting that the naïve subset depletion is associated with autoreactive antibodies in both T cell populations .
In the present study , autoreactive antibodies to platelet glycoproteins , phospholipids , dsDNA and surface antigens on T cells were observed in a subset of SIV-infected macaques that showed primary depletion of naïve CD4+ T cells . These antibodies were associated with pathologic consequences , such as thrombocytopenia , thrombosis and naïve cell depletion of both CD4+ and CD8+ T cells in blood and tissues . Our data are consistent with a potential role of these antibodies in CD4+ T cell depletion in SIV infection and has additional implications for the pathogenesis of AIDS in humans . The defining feature of HIV-associated disease is the slow loss of CD4+ T cells in the chronic phase of infection and resulting immunodeficiency [1]–[3] . However , a heightened state of generalized immune activation of T [1] , [19] , [20] and B cells [21] , [22] is also a prominent feature of HIV-infection . Indeed , the extent of immune activation in HIV-infection is an independent indicator of disease progression , as informative as plasma viral load [20] and CD4+ T cell counts . Moreover , the extent of immune activation is a striking difference between pathogenic and nonpathogenic infections with SIV [23] , [24] . The exact mechanisms whereby immune activation contributes to AIDS progression have not been clearly defined . Immune activation is presumed to lead to CD4+ T cell depletion through indirect mechanisms such as providing new activated CD4+ T cell targets and activation-induced apoptosis . Our study provides evidence that the destruction of CD4+ T cells by autoreactive antibodies may be a third mechanism that also exerts its action through immune activation . Indeed , the extent of immune activation in HIV/SIV infection has been strongly associated with the development of autoantibodies and various autoimmune diseases [31] , [32] , [38] . One such study observed a correlation with autoantibodies to nuclear , smooth muscle , and thyroid antigens with a lower CD4+ T cell count and increased mortality , consistent with prognostic value [38] . Antibodies to an array of self-antigens and resultant autoimmune disease manifestations are frequently observed in HIV infection [31] , [32] , [38] . Primary clinical manifestations which have been observed in HIV-infection include thrombocytopenia , anemia , systemic vasculitis , Reiter's syndrome , polyarthritis , Sjogren's syndrome and antiphospholipid syndrome ( APS ) [32] . Autoantibodies and autoimmune diseases have also been demonstrated in SIV-infected macaques [33] , [34] , [39] . In the present study , we observed an association of anti-platelet and antiphospholipid antibodies with clinical autoimmune disease , specifically thrombocytopenia and APS in macaques . These results suggest that the production of these antibodies had functional and clinical consequences in these macaques . The mechanisms proposed for the generation of autoreactive antibodies in HIV/SIV infection include the loss of regulatory T cells [40] , production of anti-idiotype antibodies [41] , and molecular mimicry of gp120 or gp41 with self antigens [42] , [43] . All of these implicate immune activation , especially B lymphocyte hyperactivation with hypergammaglobulinemia [21] , [22] . The hypothesis that autoimmune mechanisms play a role in AIDS pathogenesis is not a new idea . Anti-CD4+ T cell antibodies were reported soon after identification of AIDS , and a role in CD4+ T cell decline was suggested [44]–[46] . This avenue of research has not been adequately studied primarily since the field's mechanistic focus shifted to the direct effects of the virus on destruction of CCR5+CD4+ T cells [7]–[12] and the indirect effects of immune activation [15]–[18] . Additionally , although the presence of these antibodies is not at issue , it has been difficult to clearly show cause and effect . Several studies have shown an association between anti-CD4+ T cell antibodies with CD4 T+ cell depletion and disease progression [40] , [47]–[49] , suggesting the potential contribution to AIDS pathogenesis . However , none of these studies have clearly identified the mechanism within , and the subset in CD4+ T cells that is depleted . Our study demonstrated an association of anti-CD4+ T cell antibodies with loss of naïve CD4+ T cells , which is responsible for the progressive CD4+ T cell decline during chronic SIV/HIV infection . This suggests that the induction of anti-CD4+ T cell antibodies is one of the important factors causing a slow depletion of CD4+ T cells over the long asymptomatic period leading to AIDS . This mechanism probably dominates during the earlier and mid stages of HIV infection when the immune system is highly dysregulated but prior to the onset of severe immunodeficiency . Potential mechanisms by which antibodies could mediate CD4+ T cell loss including complement-mediated lysis , phagocytosis or antibody-dependent cell-mediated cytotoxicity [45] , [47] , [50] will require evaluation in these animals . In addition , autoantibodies may affect the production and differentiation of naïve T cells in thymus and bone marrow [4] , [51] , [52] . Since autoantibodies were specifcially associated with the loss of naïve CD4+ T cells rather than memory cell , this represents a distinct mechanism from the recently reported decline of the central CD4+ T cell pool , which is also an important subset for maintaining CD4+ T cells [29] , [53] . The decline of naïve cells of both CD4+ and CD8+ T cells in this study and HIV-1-infected patients [1] , [2] also suggests that the mechanism for the decline and the self antigens involved may be common to both T cells . Another autoimmune mechanism not evaluated in the present study , was the generation of autoreactive CD8+ T cells primed by the release of protein fragments from apoptotic CD4+ T cells [54] . MHC alleles may be critical to the induction of autoreactive antibodies , as indicated in a number of autoimmune diseases [55] but were not assessed in this study . Although further studies are required to define the contribution of autoantibodies to AIDS pathogenesis , our study suggests that such antibodies may act in concert with other mechanisms of CD4+ T cell loss , such as immune activation induced cell death and destruction of the lymphoid architecture [15]–[18] . Our data are consistent with a potential role of these antibodies in CD4+ T cell depletion in SIV infection and has additional implications for the pathogenesis and treatment of AIDS in humans .
Rhesus macaques ( Macaca mulatta ) of Indian origin were inoculated with molecularly cloned virus stocks generated by transfection of 293T cells , ( SIVsmE543-3 , SIVsmH635FC , SIVmac239 ) , a combination of SIVsmE543-3 and SIVsmH635FC , or terminal plasma samples from two SIVsmH445-infected macaques , H631 and H635 [6] , [26]–[28] . Twenty SIV-infected macaques used in this study consist of the 9 macaques from Kuwata et al . [27] , 8 macaques from Nishimura et al . [6] , and 3 macaques infected with plasma from SIVsm-infected macaques . Our criteria for inclusion in the study were lack of prior treatment or vaccination and the availability of terminal blood samples for analysis . EDTA-anti-coagulated blood samples were collected sequentially to examine plasma viral RNA load and lymphocyte subsets . Tissue samples , such as jejunum , ileum , colon , lung , PLN and spleen , were obtained at death from a subset of the animals . Lymphocytes were isolated from the blood , spleen and PLN by Ficoll gradient , and by collagenase treatment from jejunum , ileum , colon and lung [27] . The viral RNA loads in plasma were determined by real-time reverse transcriptase-PCR ( RT-PCR ) using a Prism 7700 sequence detection system ( Applied Biosystems , Foster City , CA ) , as previously described [27] . All animals were housed in accordance with American Association for Accreditation of Laboratory Animal Care standards . The investigators adhered to the Guide for the Care and Use of Laboratory Animals prepared by the Committee on Care and Use of Laboratory Animals of the Institute of Laboratory Resources , National Resource Council , and the NIAID Animal Care and Use Committee-approved protocols . Co-receptor usage of viruses infecting ND macaques and parental SIV strains was analyzed using CCR5 and CXCR4 antagonists , TAK779 and AMD3100 [56]–[60] . The Env gene was amplified from plasma samples of H704 and H709 at 20 wpi , and cloned into the expression vector , pcDNA3 . 1/V5-His TOPO ( Invitrogen , Carlsbad , CA ) , as described [26] . Pseudotyped viruses were prepared by cotransfection of 293T cells with pSG3ΔEnv , Rev expression plasmid and Env expression plasmid using Lipofectamine 2000 ( Invitrogen ) [61] , [62] . After a 1 h incubation of 40 μl of 2 . 5×106 TZM-bl cells [62]–[64] in the absence or presence of 10 μl of designated concentration of antagonist in 96 well-plate , 45 μl of pseudotyped viruse and 5 μl of 1 . 875 μg/ml DEAE dextran were added . After over night incubation , 100 μl fresh medium was added to wells . Luminesence was measured after 2 days using Luciferase Assay System ( Promega , Madison , WI ) and Mithras microplate luminometer ( Berthold Technologies , Bad Wildbad , Germany ) . Infection was performed in triplicate and data expressed as the mean . The following reagents were obtained through the NIH AIDS Research and Reference Reagent Program , Division of AIDS , NIAID , NIH: TAK779 , AMD3100 , pSG3ΔEnv and TZM-bl cells . Phenotypic analysis of T cells was performed by flow cytometry ( FACSCalibur; BD Biosciences , Franklin Lakes , NJ ) using FITC , PE , PerCP-Cy5 . 5 or APC-conjugated monoclonal antibodies against CD3 , CD4 , CD8 , CD28 and CD95 ( BD Biosciences ) , as described [6] , [27] . Percentages of naive and memory subsets in CD4+ or CD8+ T cells were determined using CD28 and CD95 . Live cells were gated using 7-AAD in the analyses of viably-frozen cells . Immunoglobulin on the surface of cells was detected by PE-conjugated antibodies against human IgG or human IgM with a mouse IgG1 isotype-matched control ( BD Biosciences ) . Since histograms of a control and experimental samples were overlapping , the percentages of IgG+ or IgM+ cells were determined by the Overton cumulative histogram subtraction , using mouse IgG1 isotype control [65] . To detect antibodies to T cells in the plasma , PBMC from SIV-uninfected rhesus macaques ( 5×105 cells in 50 μl ) were incubated for 1 hr at 4°C with 50 μl of plasma ( 1:5 dilution with PBS ) or purified IgG ( 1 . 5 mg/ml ) from SIV-infected macaques , washed twice with PBS supplemented with 0 . 1% bovine serum albumin , and stained with antibodies against monkey IgG ( FITC ) , CD3 ( PerCp-Cy5 . 5 ) and CD4 ( APC ) . FITC-conjugated goat anti-monkey IgG ( MP Biomedicals , Solon , OH ) was used to detect antibodies in plasma . IgG was purified from plasma samples by Nab Protein G Spin Kit ( Thermo Fisher Scientific , Rockford , IL ) . The percentages of IgG+ cells in CD4+ T cells were determined by the Overton cumulative histogram subtraction using untreated controls . Data analysis was performed using Flowjo ( TreeStar , San Carlos , CA ) . Anti-SIV antibody titers were determined by enzyme immunoassay ( EIA ) , Genetic Systems HIV2 EIA ( Bio-Rad Laboratories , Hercules , CA ) using serial dilution of plasma samples . Antibodies against three glycoproteins on platelet , GPIIb/IIIa , GPIa/IIa and GPIb/IX , were determined by EIA , PAKAUTO ( GTI Diagnostics , Waukesha , WI ) . Antibodies ( IgG ) against phopholipid in plasma samples were quantified by anti-phopholipid screen IgG/IgM ( ORGENTEC Diagnostika GmbH , Mainz , Germany ) , which detects antibodies against cardiolipin , phosphatidyl inositol , phosphatidic acid and ß2-glycoprotein I . Antibodies ( IgG ) against double-stranded DNA were quantified by anti-dsDNA ( ORGENTEC Diagnostika GmbH ) . Plasma levels of LPS were measured by the Chromogenic end point Limulus amebocyte kit ( Cambrex , Charles City , IA ) as recommended by the manufacturer . A commercially available ELISA was used to quantify plasma levels of sCD14 ( R&D Systems , Minneapolis , MN ) ; each ELISA was performed in duplicate according to the manufacturer's protocol . Cell sorting was accomplished using a FACS Aria cell sorter ( BDIS ) at 70 lb/in2 . FITC , PE , Cy5 PE , and APC were used as the fluorophores . At least 10 , 000 cells were sorted for PCR and reverse transcriptase PCR analysis . Sorted populations were consistently at least 99 . 8% pure . Quantification of SIV gag DNA in sorted CD4+ T cells was performed by quantitative PCR by means of a 5′ nuclease ( TaqMan ) assay with an ABI7700 system ( Applied Biosystems ) as previously described [66] . To quantify cell number in each reaction , qPCR was performed simultaneously for albumin gene copy number as previously described [67] . Standards were constructed for absolute quantification of gag and albumin copy number and were validated with sequential dilutions of 8E5 cell lysates that contain one copy of gag per cell . Duplicate reactions were run and template copies calculated using ABI7700 software . Ki-67 expression was assessed in formalin-fixed , paraffin-embedded lymph node samples by immunohistochemical staining as previously described [68] . Samples were incubated with monoclonal antibody against Ki-67 ( Dako , Glostrup , Denmark ) and goat anti-rabbit immunoglobulin G-Alexa fluor 488 ( Invitrogen ) . The stained sections were rinsed , counterstained with hematoxylin , covered , and photographed with a Zeiss Axiophot microscope . All statistical analyses were performed with Prizm ( GraphPad Software , La Jolla , CA ) . ND and MD macaques were compared by nonparametric Mann-Whitney U test . Correlation analysis and linear regression analysis were performed between IgG+ percentage and log naïve/memory ratio or the naïve CD4+ T cell count . | Despite intensive study , the mechanisms of CD4+ T cell depletion in human immunodeficiency virus ( HIV ) infection remain elusive . The identification of the CD4 receptor as the primary receptor for HIV seemed initially to explain the loss of CD4+ T cell lymphocytes in AIDS . However , the number of infected cells at any point in time is insufficient to explain the extent of loss and the slow disease course . While virus-induced cell death might explain early loss of CD4+ T cells , such mechanisms were unlikely to explain the slow depletion over the long period leading to AIDS . Thus , mechanisms related to the intense immune activation , also a hallmark of HIV infection , were proposed . Here , we studied a surrogate model for AIDS , simian immunodeficiency ( SIV ) infection of rhesus macaques , to explore the mechanisms of this second phase of depletion . We found that animals that exhibited slow , progressive loss of CD4+ T cells had circulating antibodies that reacted with CD4+ T cells and that the levels of these antibodies correlated with the extent of CD4+ T cell depletion . These results suggest that autoimmune destruction of CD4+ T cells represents a valid mechanism to explore in the pathogenesis of CD4+ T cell loss in AIDS . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"virology/immunodeficiency",
"viruses",
"virology/virulence",
"factors",
"and",
"mechanisms",
"virology/animal",
"models",
"of",
"infection",
"virology/host",
"antiviral",
"responses"
] | 2009 | Association of Progressive CD4+ T Cell Decline in SIV Infection with the Induction of Autoreactive Antibodies |
Avian lungs are remarkably different from mammalian lungs in that air flows unidirectionally through rigid tubes in which gas exchange occurs . Experimental observations have been able to determine the pattern of gas flow in the respiratory system , but understanding how the flow pattern is generated and determining the factors contributing to the observed dynamics remains elusive . It has been hypothesized that the unidirectional flow is due to aerodynamic valving during inspiration and expiration , resulting from the anatomical structure and the fluid dynamics involved , however , theoretical studies to back up this hypothesis are lacking . We have constructed a novel mathematical model of the airflow in the avian respiratory system that can produce unidirectional flow which is robust to changes in model parameters , breathing frequency and breathing amplitude . The model consists of two piecewise linear ordinary differential equations with lumped parameters and discontinuous , flow-dependent resistances that mimic the experimental observations . Using dynamical systems techniques and numerical analysis , we show that unidirectional flow can be produced by either effective inspiratory or effective expiratory valving , but that both inspiratory and expiratory valving are required to produce the high efficiencies of flows observed in avian lungs . We further show that the efficacy of the inspiratory and expiratory valving depends on airsac compliances and airflow resistances that may not be located in the immediate area of the valving . Our model provides additional novel insights; for example , we show that physiologically realistic resistance values lead to efficiencies that are close to maximum , and that when the relative lumped compliances of the caudal and cranial airsacs vary , it affects the timing of the airflow across the gas exchange area . These and other insights obtained by our study significantly enhance our understanding of the operation of the avian respiratory system .
The anatomical structure and airflow dynamics of the avian respiratory system are remarkably different to that of mammalian lungs [1] . The anatomical structure is complex , with multiple flexible airsacs that act like bellows to ventilate rigid tubes ( parabronchi ) in which gas exchange occurs , and a complicated branching structure that produces aerodynamic valving [1 , 2] . The airflow through the parabronchi ( lungs ) is unidirectional; flowing from the caudal ( back ) group of airsacs to the cranial ( front ) group of airsacs during both inspiration and expiration . ( More precisely , the flow is unidirectional through the paleopulmonic-parabronchi that lie between the caudal and cranial airsacs . Some birds also contain neopulmonic-parabronchi in which airflow is bidirectional , but it forms a small part of the gas exchange surface area—less than 30% [2] . In this paper we use the term parabronchi to refer to the paleopulmonic parabronchi unless otherwise indicated . ) Unlike in the mammalian respiratory system , the functions of ventilation and gas exchange have been uncoupled in the avian respiratory system; specifically , the flow of air through the system is caused by large flexible airsacs , whilst gas exchange occurs in narrow parabronchi which are rigid and firmly bound to the ribs [2] . The narrow , rigid structure of the parabronchi is thought to be related to the finding that birds have a thinner but mechanically stronger blood-gas barrier than equivalent mammals [3 , 4] . Furthermore the structure of the parabronchi and blood capillaries allows for cross-current gas exchange . These features are thought to contribute to the increased gas exchange efficiency of birds compared to mammals , especially at high-altitude or in a hypoxic environment [3 , 5–8] . The airflow pattern within the avian respiratory system is widely agreed upon . It has been determined by direct measurements of flow rates [9–11] , as well as by experiments that used tracer gas , or CO2 and O2 measurements to indirectly determine the flow [10 , 12–15] . An important factor leading to unidirectional flow is hypothesized to be the effective inspiratory and expiratory aerodynamic valving that results from the interaction between the complex anatomical structure , including airway branching and constrictions , and the fluid dynamics involved [1 , 2] . The relative importance of the two valves has not been investigated . Additionally , the complex branching structure within the system affects the resistance to airflow of the different sections of the system . The effect of these resistances and the importance of their relative differences in generating the flow pattern is not known . Recently unidirectional airflow has been found in the lungs of some reptiles ( specifically alligators [16 , 17] , crocodiles [18 , 19] , iguanas [20] , and monitor lizards [21] ) . Comparing avian and reptile systems , which have very different levels of anatomical complexity in terms of the branching structures and the presence or absence of airsac separation , will provide important insights into the aerodynamic valving in both birds and reptiles . Mathematical modelling of the avian respiratory system has focused mainly on the gas exchange in the parabronchi ( for example , [6 , 22–28] ) . These studies determined that gas exchange is cross-current and found gas exchange parameters for a range of avian species and experimental conditions [1 , 2] . Existing mathematical models of the airflow through the avian system had limited success in producing unidirectional flow [29 , 30] . Urushikubo et al . [30] used a three dimensional spatial model with simplified geometry for the pathways within the respiratory system , coupled with flexible airsacs . They found unidirectional flow through the parabronchi , but only for some parameter values . Additionally , the flow did not show any inspiratory or expiratory valving . Maina et al . [29] investigated aerodynamic inspiratory valving in ostriches by constructing a three dimensional anatomical model of the junction between the ventrobronchial branches and the main mesobronchus ( i . e . the junction between the airways that lead air to the caudal airsacs and the airways that lead air from the cranial airsacs ) . Using computational fluid dynamics simulations they were only able to reproduce inspiratory valving if they included additional branches downstream ( the secondary dorsobronchi branches ) , showing the importance of including the whole system when investigating aerodynamic valving . Unidirectional flow exists in all birds , despite massive inter-species differences in anatomy , and across most experimental conditions—including when ventilating the respiratory system post-mortem [31] . Thus , a useful mathematical model of the airflow in the avian respiratory system must produce unidirectional flow through the parabronchi across a broad range of parameter values and frequencies . In this paper we present a new , relatively simple , mathematical model of avian respiration that reproduces the airflow pattern described above . The unidirectional flow in our model is robust to changes in frequency and model parameters , and has efficiencies , flow rates , and pressures that match experimental findings . Additionally , our model generates several novel insights on the role of inspiratory and expiratory valving and the importance of variations in the airflow resistances and airsac compliances within the system that are thought to occur during respiration and in response to stimuli including hypoxia ( lack of oxygen ) and hypercapnia ( excess of carbon dioxide ) . We first describe the mathematical model and then the new insights it produced . The model development and mathematical analysis are described later in the Methods section .
A schematic model of the avian respiratory system is shown in Fig 1 . For simplicity , only one side of the respiratory system is shown ( see Fig 12 for the full model ) . The caudal and cranial airsacs are considered to be flexible with lumped compliances C1 and C2 respectively and averaged pressures P1 and P2 respectively . The pressure in the coelom ( thoracic-abdominal cavity ) outside both sets of airsacs , Pext ( t ) , varies periodically due to the respiratory muscles , which causes the airsacs to inflate and deflate . During inspiration ( indicated by blue solid arrows in Fig 1 ) , air flows in through the beak along the trachea ( qT ) , through the primary and meso-bronchi to the caudal airsacs ( q1 ) , and from the caudal airsacs to the cranial airsacs through the parabronchi ( qP ) . During expiration ( indicated by green dashed arrows in Fig 1 ) , air flows from the caudal airsacs to the cranial airsacs through the parabronchi ( qP ) , from the cranial airsacs through the ventrobronchi ( q2 ) and along the trachea to exit the beak ( qT ) . The airflow pathways are considered to be rigid ( no compliance ) and to have resistance to airflow , Ri , where i ∈ {1 , 2 , T , P} . Note that since our aim is to create a model that is applicable generally across all birds , and we are primarily interested in understanding the unidirectional flow through the paleopulmonic-parabronchi , we have chosen to include only the paleopulmonic-parabronchi explicitly . However , the neopulmonic-parabronchi are included indirectly in our model through the lumped resistance parameters . From the pressures P1 , P2 and Patm ( atmospheric pressure ) and the resistances , Ri , we can calculate the airflow , qi , through every pathway in the system ( Eqs ( 17 ) – ( 20 ) ) . Using the relationship between the pressure , compliance , and volume , assuming that the compression of air is negligible , and applying some algebraic manipulations ( refer to the Methods section for more details ) , we get the following two equations for the rate of change of the pressures P1 and P2: d P 1 d t = - R P R 2 ( P 1 - P a t m ) - ( R P R T + R ¯ ) ( P 1 - P 2 ) C 1 R ¯ R P + d P e x t d t ( 1 ) d P 2 d t = - R P R 1 ( P 2 - P a t m ) + ( R P R T + R ¯ ) ( P 1 - P 2 ) C 2 R ¯ R P + d P e x t d t ( 2 ) where R ¯ = R 1 R 2 + R 2 R T + R T R 1 . The resistances R1 and R2 are discontinuous and vary depending on the flow direction as shown in Fig 2 . This allows us to produce effective inspiratory and expiratory valving: A representative example of the results found in this model is shown in Fig 3 , with the parameter values listed in Table 1 . In Fig 3A we see that the pressure differences between the airsacs and atmospheric pressure ( x1 = P1 − Patm and x2 = P2 − Patm ) are orders of magnitude greater than the difference between the two airsac pressures ( x1 − x2 ) . This matches experimental measurements , e . g . [12 , 15] . Fig 3B shows that the flow through the parabronchi is unidirectional ( qP > 0 ) and that the valving is working well as q2 ≈ 0 during inspiration and q1 ≈ 0 during expiration . The tidal volume is 36 . 0 mL and the combined flow through the parabronchi per breath is 31 . 3 mL on both sides , so most of the air which is breathed in passes through the gas exchange area . Fig 3C shows the volumes of the caudal set of airsacs , V1 , and the cranial set of airsacs , V2 , on one side of the respiratory system ( see also Fig 12 ) . The volumes can be calculated directly from Eqs ( 9 ) and ( 10 ) . The ventilation volume into each set of airsacs is calculated using max ( Vi ) -min ( Vi ) for i = 1 , 2 and is independent of the chosen parameters Pc , and Vi , res . For the results shown in Fig 3 the ventilation volume is found to be 21 . 5 mL per breath in total for the caudal airsacs , on both sides , and 16 . 3 mL per breath in total for the cranial airsacs , on both sides . These values match experimental data [38] . Note that the sum of the ventilation of all the airsacs can be greater than or less than the tidal volume , as some air flows past the airsacs , and some air flows into both sets of airsacs . By analysing the phase plane dynamics of the system of Eq ( 1 ) ( see Methods section ) , we can show that airflow through the parabronchi will be unidirectional ( qP > 0 ) when γR1 ≤ R2 during inspiration and γR1 ≥ R2 during expiration , where γ = C1/C2 . In the borderline case , where γR1 = R2 during both inspiration and expiration , the flow through the parabronchi , qP , is zero . A combination of effective inspiratory and expiratory valving ( γR1 < R2 during inspiration and γR1 > R2 during expiration ) will produce unidirectional flow . However , unidirectional flow could also be achieved by inspiratory or expiratory valving alone , e . g . effective inspiratory valving , where γR1 < R2 during inspiration and γR1 = R2 during expiration , or effective expiratory valving , where γR1 = R2 during inspiration and γR1 > R2 during expiration . Fig 4 shows a sketch of the phase plane dynamics in the case of effective inspiratory valving , where the variables are transformed such that x1 = P1 − Patm and x2 = P2 − Patm . The stable equilibrium P1 = P2 = Patm in the absence of pressure variations ( no breathing ) is then at the origin ( 0 , 0 ) . The line x2 = x1 marks all the possible pressures for which the flow qP is zero . Above this line ( x2 > x1 ) qP is negative ( marked in shaded grey ) , and below it ( x2 < x1 ) qP is positive . The dark blue curves show the solutions to the system from different initial conditions when there is no breathing , and thus no change in the external pressure . All these solutions approach the origin ( 0 , 0 ) . The red curve shows the solution ( not to scale ) of the system when the external pressure Pext is changing due to breathing . This change in pressure is the same outside both sets of airsacs and thus acts along the vector [1 , 1]T . It can be seen that the system is ‘trapped’ below the line x1 = x2 and therefore the airflow is unidirectional ( see Methods section for more details ) . In Fig 4 inspiration is marked by the light blue region and expiration is marked by the green region . When both P1 and P2 are greater than Patm ( upper right quadrant ) it is obvious that expiration will occur ( qT < 0 ) . When both P1 and P2 are less than Patm ( lower left quadrant ) , inspiration will occur ( qT > 0 ) . The exact place ( in the lower right quadrant ) where the transition from inspiration to expiration occurs in the phase plane will depend on the chosen parameter values . Specifically , the flow qT will change direction on the line x 2 = - R 2 , i n s p R 1 , e x p x 1 with inspiration occurring if x 2 < - R 2 , i n s p R 1 , e x p x 1 and expiration occurring if x 2 > - R 2 , i n s p R 1 , e x p x 1 ( see Methods section and Fig 13 ) . For simplicity , in Fig 4 we consider the case where R2 , insp = R1 , exp and the transition between inspiration and expiration occurs on the line x2 = −x1 . The conditions for unidirectional flow do not depend on the frequency or amplitude of breathing . Consequently , unidirectional flow will persist as long as the conditions on the resistances and compliances ( stated in the previous section ) are satisfied . Increasing the amplitude , Pamp , increases the flow rates proportionally ( see Fig 5 ) . As the period , T , decreases ( the frequency increases ) the mean flow through the parabronchi remains relatively constant , but the variation in the flow rate during the breathing cycle decreases and the flow becomes more constant ( see Fig 6 ) . This matches what is seen experimentally [11] , and may have an impact on the efficacy of gas exchange . In the avian system , the aerodynamic valving is not 100% effective . Inspiratory valving is found experimentally to be 95–100% effective [14 , 32–34 , 39] . However , expiratory valving efficacy varies greatly between species and experimental conditions: 76–90% in chickens [12 , 40] , 88% in ducks [14] , and 95% in geese [37] , with a strong dependence on gas velocity; at higher flow rates ( exercise conditions ) the valve is more effective than at rest . When fresh air flows into the cranial or caudal airsacs and is then breathed back out without passing through the parabronchi , this air does not undergo gas exchange , and is thus wasted . We define the efficiency of the whole system as the fraction of the tidal volume that passes through the parabronchi . For example , when efficiency = 1 all the air that is inhaled will pass through the parabronchi and undergo gas exchange . We calculate the efficiency in our model by numerically integrating the airflow qP during one cycle of breathing to find the total volume of air that flows through the parabronchi per breath , and numerically integrating the flow qT during inspiration to calculate the tidal volume ( the total air inhaled per breath ) . The ratio of volume through parabronchi per breath to tidal volume gives us a measure of how efficient the lung system is . Efficiency = ∫ INSP + EXP q P dt ∫ INSP q T dt ( 5 ) where ∫INSP indicates the definite integral during inspiration and ∫INSP+EXP indicates the definite integral during one breath . If the model includes effective inspiratory valving only ( R2 , insp ≫ γR1 , insp and R2 , exp = γR1 , exp ) with R1 , insp = R1 , exp , the flow qP is unidirectional ( qP > 0 ) , but the maximum efficiency we can find numerically is around 50% . The cause of this low efficiency is that a large proportion of the fresh air that flows into the caudal airsacs , then flows back out ( q1 < 0 ) without participating in gas exchange , as shown in Fig 7A . In Fig 7A , the inspiratory valving efficacy = 98 . 7% , the expiratory valving efficacy = 47 . 7% , the overall efficiency is 47 . 1% , the tidal volume is 38 . 1 mL whilst the flow through both parabronchi per breath is only 17 . 9 mL . Similarly , if the model includes only effective expiratory valving ( γR1 , insp = R2 , insp and γR1 , exp ≫ R2 , exp ) with R2 , insp = R2 , exp , we find numerically that the maximum efficiency we can reach is around 50% , due to flow into the cranial airsacs during inspiration ( q2 < 0 ) . An example of effective expiratory valving is shown Fig 7B , where the inspiratory valving efficacy is 48 . 6% and the expiratory valving efficacy is 87 . 0% . This gives an overall efficiency of 42 . 3%; from a tidal volume of 38 . 4 mL and only 16 . 2 mL flow through both parabronchi per breath . By including both inspiratory and expiratory valving we find that we can reduce this back flow , and when we match the experimental valving efficiencies of 98–100% for inspiratory valving and ≈88% for expiratory valving , we find that the overall efficiency of the system matches those found experimentally , whilst maintaining realistic resistance and compliance values . The flow rates for our chosen parameter values are shown in Fig 3B , where the inspiratory valving efficacy is 98 . 0% , the expiratory valving efficacy is 88 . 6% , and the overall efficiency is 86 . 8% . In our model , we can investigate the impact of varying the resistances R1 , insp and R2 , exp . We need to keep R1 , insp + R2 , exp constant so that there is no change in the total resistance of the system , here we choose R1 , insp + R2 , exp = 6 cmH2O/L⋅s . Additionally , it is important to keep R2 , insp = 100 × R1 , insp and R1 , exp = 10 × R2 , exp , so that the strength of the valving isn’t changing . Fig 8 shows that depending on the ratio of compliances , γ , the maximum efficiency will be for 0 . 2 < R1 , insp/R2 , exp < 1 . This is consistent with experimental observations that found R1 , insp to be lower than R2 , exp ( see Methods section ) . Comparatively , we find that varying γ and Ctot affect the efficiency by less than 1% ( see Selecting model parameters ) . The airflow through the parabronchi is not constant during the breathing cycle . An important feature of the flow through the parabronchi is that it can be observed to occur mostly during inspiration , or expiration , or both , depending on parameter values and experimental conditions [10 , 11] . From Fig 7 we can see that the aerodynamic valving affects the timing of the airflow through the parabronchi; effective inspiratory valving increases parabronchial flow during inspiration ( Fig 7A ) , whereas effective expiratory valving increases parabronchial flow during expiration ( Fig 7B ) . However , once we fix the valving efficacy to physiologically realistic levels ( e . g . inspiratory valving 98% , expiratory valving 88% ) , the ratio of volume flowing through the parabronchi during inspiration and expiration ( I:E parabronchial volume flow ratio ) due to the valving is fixed . We calculate the ratio of volume flowing through the parabronchi during inspiration and expiration from the output of our model as follows: I : E parabronchial volume flow ratio = ∫ INSP q P dt ∫ EXP q P dt ( 6 ) For the default parameter values ( Table 1 ) the I:E parabronchial volume flow ratio is 0 . 862 , i . e . there is slightly less flow during inspiration than during expiration , as shown in Fig 3 . Varying γ has a major impact on the timing of the flow through the parabronchi ( recall that γ = C1/C2 ) . When γ is low the majority of the flow through the parabronchi occurs during inspiration , while when γ is high the flow through the parabronchi occurs mostly during expiration . In Fig 9 we plot the I:E parabronchial volume flow ratio as a function of γ . As γ increases we find that the majority of the flow qP moves from being during the inspiratory phase to being during the expiratory phase . This result is conserved for a range of total compliance ( C1 + C2 = Ctot ) values . Despite this change in the timing of the flow , the system’s overall efficiency only decreases slightly ( from 87 . 7% to 86 . 4% ) when γ increases ( see S1A Fig ) , and the airflow through the parabronchi remains unidirectional . These results are consistent with experimental observations . Anatomically , the caudal and cranial airsacs are found to have different properties [2] . In ducks , Scheid et al . [38] found that the caudal airsacs are more compliant and have larger ventilation volume changes than the cranial airsacs , especially during relaxed ( anaesthetized ) breathing . Furthermore , the ratio of compliances varies between individuals and species [1] and many variations in the flow pattern are also observed [10 , 11] . For example , in spontaneously breathing geese the flow through the parabronchi increases at the end of inspiration and peaks during expiration [10] . Looking at ducks it is found that the flow during inspiration is higher than the flow during expiration when panting , the flow during inspiration and expiration are similar in spontaneous breathing , and when relaxed ( anaesthetized ) the flow rate is much stronger during expiration [11] . Our results as we vary γ provides similar changes in flow patterns ( Fig 10 ) , which agree with experimental findings; γ should decrease during exercise when the abdominal and chest muscles are stiff , and increase under relaxation conditions when the muscles relax . We find that changing the relative resistances R1 , insp/R2 , exp does affect the timing of the flow through the parabronchi slightly , with more flow during expiration as R1 , insp/R2 , exp increases ( see S2A Fig ) . We also find that varying the total compliance Ctot does not change the timing of the flow through the parabronchi substantially , but the strength of the effect decreases at high Ctot ( see S2B Fig ) . Overall , the ratio of compliances , γ , is the dominant effect . We observe that although the forcing of the system is symmetric ( sinusoidal function ) , the duration of inspiration , Ti ( measured as the time during which qT > 0 ) , is not always equal to the duration of expiration , Te ( measured as the time during which qT < 0 ) . For our chosen default parameter values ( Table 1 ) , with period T = 3s , Ti = 1 . 4s and Te = 1 . 6s , and the ratio of inspiration duration to expiration duration ( I:E time ratio = Ti/Te ) is 0 . 89 . This asymmetry varies depending on parameter values . We investigate the impact of varying the resistances R1 , insp and R2 , exp , while the overall resistance of the system and the strength of the valving constant , as before . We find that when we decrease R1 , insp relative to R2 , exp the duration of expiration increases , with a concordant decrease in the duration of inspiration ( Fig 11 ) . The ratio of compliances , γ , and the total compliance , Ctot , do affect the I:E time ratio slightly , but the impact of varying the relative resistances is much stronger ( see Selecting model parameters ) .
The complex anatomical structure of the avian respiratory system has been represented in our model by discontinuous resistances ( R1 and R2 ) that depend on the direction of airflow through them . These resistance values could depend on the properties of the flow and would then vary with frequency and amplitude of breathing , as well as with other parameters such as muscle tone that we did not take into account in our model . Nevertheless , we have shown that as long as γR1 ≤ R2 during inspiration and γR1 ≥ R2 during expiration , where γ = C1/C2 is the ratio of the airsac compliances , unidirectional flow will persist . We also assumed that both the inspiratory and expiratory valving are highly effective which is true during regular breathing but may not be the case if breathing consists of very high or very low frequencies or amplitudes . In particular , we note that experimentally it is found that panting and other breathing patterns ( bird song/calls ) do not have the same pattern . For example , during panting the expiratory valving is not strong and there is a large amount of air shunted into the primary bronchus ( q1 < 0 ) which bypasses the parabronchi [9] . The two discontinuous resistances in our model make the system nonlinear , despite the assumptions of constant resistance and compliance elements . The presence of discontinuities in models is known to produce complicated phenomena , especially in non-autonomous systems with external forcing [41] . In this model , we have found the intriguing result that the inspiratory and expiratory periods are uneven in response to regular ( sinusiodal ) forcing . The phenomenon underlying this disparity is not clear yet . Further theoretical analysis is left for future investigations . In summary , the new mathematical model we have developed , and the analytical and computational study we have conducted , significantly increase our understanding of unidirectional airflow in avian lungs . Our new model is broadly applicable across all birds and can be extended or integrated into larger systems-level studies of the avian respiratory system . Our model also provides a new example of a non-smooth dynamical system and will be used in future investigations of the human respiratory system through comparative physiology .
Fig 12 shows the full model including left and right sides of the respiratory system . The caudal and cranial airsacs have pressures P1 and P2 respectively , and compliances C1 and C2 respectively . All other pathways and junctions are assumed to be rigid . To simplify our analysis , we assume that the left and right sides of the bird are symmetrical; this enables us to reduce the model to the system shown in Fig 1 , where RT = 2Rtrachea + REPPB . In the remainder of this work we will use the model shown in Fig 1 , which considers only one side . To find the overall flow in the whole animal , we simply double the flow rates found from this single side . To construct the mathematical model we begin by calculating the rate of change of volume in the caudal and cranial airsacs ( dV1/dt and dV2/dt , respectively ) , assuming that the compression of air is negligible: d V 1 dt = q 1 - q P ( 7 ) d V 2 dt = q P - q 2 ( 8 ) where qi with i ∈ {1 , 2 , P} is the airflow through the corresponding section . Next we assume that the airsacs are elastic , with compliance C1 and C2 . Additionally , surrounding both sets of airsacs there is an external pressure Pext ( t ) , which has a time-varying component that represents the change in pressure generated by the muscles of the chest and abdomen during breathing . This gives the equations: V 1 = C 1 P 1 - P e x t + V 1 , r e s ( 9 ) V 2 = C 2 P 2 - P e x t + V 2 , r e s ( 10 ) where V1 , res and V2 , res are the resting volumes of the caudal and cranial airsacs when the pressure difference between the airsacs and the surrounding thoracic-abdominal cavity ( coelom ) is zero . In all our simulations we use the sinusoidal function: P e x t ( t ) = P c - P a m p 2 cos 2 π t T ( 11 ) to model the time-varying pressure outside the airsacs . This function oscillates with a peak-to-peak amplitude of Pamp , which is the amplitude of the forcing from breathing , around a pressure Pc , which is the baseline pressure in the coelom . Differentiating Eqs ( 9 ) and ( 10 ) with respect to time gives d V 1 dt = C 1 d P 1 dt - d P e x t dt ( 12 ) d V 2 dt = C 2 d P 2 dt - d P e x t dt ( 13 ) with d P e x t dt = P a m p π T sin 2 π t T ( 14 ) Equating Eqs ( 12 ) and ( 7 ) , and Eqs ( 13 ) and ( 8 ) gives: d P 1 dt = q 1 - q P C 1 + d P e x t dt ( 15 ) d P 2 dt = q P - q 2 C 2 + d P e x t dt ( 16 ) Assuming laminar flow , we obtain expressions for the flow rates qT , q1 , q2 , and qP in terms of our variables P1 and P2 , as well as the pressures Patm and PJ: q T = P a t m - P J R T ( 17 ) q 1 = P J - P 1 R 1 ( 18 ) q 2 = P 2 - P J R 2 ( 19 ) q P = P 1 - P 2 R P ( 20 ) From the geometry of the system , the flow at junction J is conserved ( inflexible junction ) , so qT + q2 = q1 . From this , and flow Eqs ( 17 ) – ( 19 ) we find: R T ( q 1 - q 2 ) + P J = P a t m ( 21 ) - R 1 q 1 + P J = P 1 ( 22 ) R 2 q 2 + P J = P 2 ( 23 ) Solving for q1 , q2 , and PJ in terms of P1 , P2 , and Patm we get: q 1 = - R 2 ( P 1 - P a t m ) - R T ( P 1 - P 2 ) R ¯ ( 24 ) q 2 = R 1 ( P 2 - P a t m ) - R T ( P 1 - P 2 ) R ¯ ( 25 ) P J = R 2 R T P 1 + R T R 1 P 2 + R 1 R 2 P a t m R ¯ ( 26 ) where we introduce the combined resistance R ¯ = R 1 R 2 + R 2 R T + R T R 1 to simplify the equations . Substituting the expressions for the flow rates Eqs ( 20 ) , ( 24 ) and ( 25 ) into Eqs ( 15 ) and ( 16 ) , we get the rate Eqs ( 1 ) and ( 2 ) for P1 and P2 respectively . The unique steady-state for the system of ordinary differential Eqs ( 1 ) and ( 2 ) in the absence of changing pressure due to breathing ( d P e x t dt = 0 ) , is P1 = P2 = Patm . If we move the equilibrium point to the origin using the transformation x1 = P1 − Patm , x2 = P2 − Patm , the system of equations becomes: d x 1 dt = - ( R P R T + R P R 2 + R ¯ ) C 1 R ¯ R P x 1 + ( R P R T + R ¯ ) C 1 R ¯ R P x 2 + d P e x t dt ( 27 ) d x 2 dt = ( R P R T + R ¯ ) C 2 R ¯ R P x 1 + - ( R P R T + R P R 1 + R ¯ ) C 2 R ¯ R P x 2 + d P e x t dt ( 28 ) The flow rates in terms of these new variables are: q T = - R 2 x 1 - R 1 x 2 R ¯ ( 29 ) q 1 = - R 2 x 1 - R T ( x 1 - x 2 ) R ¯ ( 30 ) q 2 = R 1 x 2 - R T ( x 1 - x 2 ) R ¯ ( 31 ) q P = x 1 - x 2 R P ( 32 ) Eqs ( 27 ) and ( 28 ) can be written in matrix form as: d X → dt = A X → + d P e x t dt 1 1 ( 33 ) where A = - ( R P R T + R P R 2 + R ¯ ) C 1 R ¯ R P ( R P R T + R ¯ ) C 1 R ¯ R P ( R P R T + R ¯ ) C 2 R ¯ R P - ( R P R T + R P R 1 + R ¯ ) C 2 R ¯ R P ( 34 ) and X → = [ x 1 , x 2 ] T . If we consider the unforced system , where d P e x t dt = 0 , Eq ( 33 ) becomes the autonomous , homogeneous , linear system d X U → dt = A X U → , where X U → = [ x 1 , x 2 ] T . The solution to this autonomous linear system is: X U → ( t ) = a 1 e λ 1 t V 1 → + a 2 e λ 2 t V 2 → ( 35 ) where λi are the eigenvalues of A , V i → are their corresponding eigenvectors , and ai depend on the specific initial conditions . To find explicit expressions for the eigenvalues and eigenvectors of the unforced system , we simplify our analysis by scaling time by C 1 R ¯ R P . This transformation does not change the phase plane dynamics of the system . The matrix A ( Eq ( 34 ) ) becomes: A ^ = - ( R 2 R P + β ) β γ β - γ ( R 1 R P + β ) ( 36 ) where β = R T R P + R ¯ , and γ = C1/C2 . The eigenvalues are then given by: λ 1 = T r ( A ^ ) - T r ( A ^ ) 2 - 4 Det ( A ^ ) 2 ( 37 ) λ 2 = T r ( A ^ ) + T r ( A ^ ) 2 - 4 Det ( A ^ ) 2 ( 38 ) where: T r ( A ^ ) = - ( γ R 1 + R 2 ) R P - ( γ + 1 ) β ( 39 ) Det ( A ^ ) = γ ( R 1 R 2 R P 2 + β ( R 1 + R 2 ) R P ) ( 40 ) These eigenvalues will be real if Δ = Tr ( A ^ ) 2 - 4 Det ( A ^ ) ⩾ 0 . In this system all resistance and compliance values must be positive . Using this simple physiological constraint we find that T r ( A ^ ) < 0 and Det ( A ^ ) > 0 for all values of Ci , Ri > 0 . Additionally , it can be shown that Δ = Tr ( A ^ ) 2 - 4 Det ( A ^ ) > 0 , and thus the system has two real eigenvalues , λ1 < λ2 < 0 , for all parameter values . The corresponding eigenvectors are: V 1 → = 1 λ 1 + R P R 2 + β β ( fast ) ( 41 ) V 2 → = 1 λ 2 + R P R 2 + β β ( slow ) ( 42 ) All solutions ( given by Eq ( 35 ) ) will tend to the single steady-state ( x1 , x2 ) = ( 0 , 0 ) as t → ∞ by initially following the fast eigenvector V 1 → and then approaching the steady-state ( more slowly ) following the slow eigenvector V 2 → . This behaviour is characteristic of solutions to linear ordinary differential equations . In our model , we are looking for solutions where qP ≥ 0 . From Eq ( 32 ) we know that qP = 0 on the line x2 = x1 and solutions above the line x2 = x1 will have qP < 0 while solutions below the line x2 = x1 will have qP > 0 . Thus , to maintain positive unidirectional flow ( qP ≥ 0 ) , solutions must lie on or below the line x2 = x1 . Putting this information together with the knowledge that the unforced system will return to steady state ( x1 , x2 ) = ( 0 , 0 ) following the slow eigenvector V 2 → , we can conclude that solutions will be pushed into the region qP < 0 if the slow eigenvector , V 2 → lies above the line x2 = x1 . Consequently , in order to maintain unidirectional flow we must find conditions that ensure that the slow eigenvector lies on or below the line x2 = x1 . From Eq ( 42 ) , V 2 → = [ 1 , 1 ] T when λ 2 + R P R 2 + β β = 1 . Rearranging this equation , we find that the slow eigenvector will lie along the line x2 = x1 when γR1 = R2 . When perturbed by breathing , Pext is applied along the vector [1 , 1]T ( see Eq ( 33 ) ) . So in the case where γR1 = R2 and V 2 → = [ 1 , 1 ] T , the system will oscillate ( due to the forcing from breathing ) along the line x2 = x1 and qP = 0 . This forms a useful boundary case . During inspiration the slow eigenvector will lie below the line x1 = x2 ( qP > 0 ) if λ 2 + R P R 2 + β β > 1 , which occurs when γR1 < R2 . During expiration , the slow eigenvector will lie below the line x1 = x2 ( qP > 0 ) if λ 2 + R P R 2 + β β < 1 , which occurs when γR1 > R2 . If both these conditions are satisfied , then the dynamics of the unforced system tells us that solutions will move quickly into the region qP > 0 and will stay there . When the system is perturbed by breathing , the forcing Pext is applied along the vector [1 , 1]T , and thus will not move the solutions into the region qP < 0 . This analysis gives the following conditions for unidirectional flow: γR1 ≤ R2 during inspiration and γR1 ≥ R2 during expiration . The conditions for unidirectional flow stated above can only be satisfied if the resistances or compliances change between inspiration and expiration . As there is no evidence that the compliances would change during a single breathing cycle , we look instead at changing the resistances . Experimental findings have shown that the complicated anatomical structure causes effective aerodynamic valving , where the flow through the different pathways differs between inspiration and expiration . To reproduce this effect in our model we make the resistances R1 and R2 dependent on flow direction . Inspiratory valving is incorporated into our model by increasing R2 when q2 < 0 ( see Fig 2 ) to reduce the negative q2 flow during inspiration . The resistance R2 is: R 2 = R 2 , exp + ( R 2 , i n s p − R 2 , exp ) H ( P J − P 2 ) where H denotes the Heaviside function , R2 , exp is the physiological value for resistance to flow in the preferred direction ( q2 > 0 ) and R2 , insp is the higher effective resistance value during inspiration due to the inspiratory valving . Expiratory valving is incorporated into our model by increasing R1 when q1 < 0 to prevent flow from the caudal airsacs during expiration ( see Fig 2 ) . The resistance R1 is: R 1 = R 1 , i n s p + ( R 1 , exp − R 1 , i n s p ) H ( P 1 − P J ) where H denotes the Heaviside function , R1 , insp is the physiological value for resistance to flow in the preferred direction ( q1 > 0 ) and R1 , exp is the higher effective resistance value during inspiration due to the expiratory valving . From Eq ( 30 ) the flow q1 = 0 when x 2 = R 2 + R T R T x 1 , for qP > 0 this line lies in the lower left quadrant of the ( x1 , x2 ) -phase plane . From Eq ( 31 ) the flow q2 = 0 when x 2 = R T R 1 + R T x 1 , for qP > 0 this line lies in the upper right quadrant of the ( x1 , x2 ) -phase plane . From Eq ( 29 ) we can show that inspiration ( qT > 0 ) occurs in the region x 2 < - R 2 R 1 x 1 , and expiration ( qT < 0 ) occurs in the region x 2 > - R 2 R 1 x 1 . Combining all this information we can sketch the flow direction transitions onto the phase plane ( Fig 13 ) , where: Note that the qT = 0 transition always occurs in the lower right quadrant , where q1 < 0 and q2 < 0 . This means that we can define inspiration as the region x 2 < - R 2 , i n s p R 1 , exp x 1 , and expiration as the region x 2 > - R 2 , i n s p R 1 , exp x 1 . Also note that the transition through regions ( 2 ) and ( 3 ) is very fast . We used Matlab’s event detection in conjunction with the solver ode23 to change R1 and R2 values when q1 and q2 crossed through zero . Starting at region ( 1 ) ( q1 > 0 , q2 < 0 ) the sequence for one breath is: This sequence was then repeated for as many breaths as required until steady state was reached . Steady state was numerically defined as being reached when the area under the curve qT was zero ( less than 1 × 10−5 ) over a single breath . That is , the flow into the bird was equal to the flow out of the bird in each breath . The volumetric flow ( area under the qi curves ) through each segment was found numerically using the trapezoidal method . In the results presented , a step size of δt = 1 × 10−4 was used . The step size was reduced in several cases to test convergence and the results were consistent . Note: We state and discuss resistances with the units cmH2O/L⋅s , but use the units mL/cmH2O for compliances and mL for volumes , based on common practice in the field . When implementing the model it is important to use consistent units ( mL or L only ) . We selected parameters to match duck respiratory systems , as we have the best data on airsac compliance and ventilation for this species . The default parameter values given in Table 1 are used for all the numerical calculations unless otherwise is indicated in the figure legends or in the text . Below we explain in more detail how we chose the specific parameters . | Birds and mammals have similar metabolic demands and cardiovascular systems , but they have evolved drastically different respiratory systems . A key difference in birds is that gas exchange occurs in rigid tubes , through which air flows unidirectionally during both inspiration and expiration . How this unidirectional flow is generated , and the factors affecting it , are not well understood . It has been hypothesized that the unidirectional flow is due to aerodynamic valving resulting from the complex anatomical structure . To test this hypothesis we have constructed a novel mathematical model that , unlike previous models , produces unidirectional flow through the lungs consistently even when the amplitude and frequency of breathing change . We have investigated the model both analytically and computationally and shown the importance of aerodynamic valving for generating strong airflow through the lungs . Our model also predicts that the timing of airflow through the lungs depends on the relative compliances of the different airsacs that exist in birds . The lumped parameters approach we use means that this model is generally applicable across all birds . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"livestock",
"medicine",
"and",
"health",
"sciences",
"dynamics",
"classical",
"mechanics",
"fluid",
"mechanics",
"vertebrates",
"animals",
"physiological",
"processes",
"respiratory",
"system",
"breathing",
"respiration",
"waterfowl",
"air",
"flow",
"birds",
"ducks",
"fluid",
"dynamics",
"fowl",
"continuum",
"mechanics",
"poultry",
"agriculture",
"fluid",
"flow",
"physics",
"flow",
"rate",
"anatomy",
"physiology",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"organisms",
"aerodynamics"
] | 2016 | Robust Unidirectional Airflow through Avian Lungs: New Insights from a Piecewise Linear Mathematical Model |
Canine rabies was reintroduced to the city of Arequipa , Peru in March 2015 . The Ministry of Health has conducted a series of mass dog vaccination campaigns to contain the outbreak , but canine rabies virus transmission continues in Arequipa’s complex urban environment , putting the city’s 1 million inhabitants at risk of infection . The proximate driver of canine rabies in Arequipa is low dog vaccination coverage . Our objectives were to qualitatively assess barriers to and facilitators of rabies vaccination during mass campaigns , and to explore strategies to increase participation in future efforts . We conducted 8 focus groups ( FG ) in urban and peri-urban communities of Mariano Melgar district; each FG included both sexes , and campaign participants and non-participants . All FG were transcribed and then coded independently by two coders . Results were summarized using the Social Ecological Model . At the individual level , participants described not knowing enough about rabies and vaccination campaigns , mistrusting the campaign , and being unable to handle their dogs , particularly in peri-urban vs . urban areas . At the interpersonal level , we detected some social pressure to vaccinate dogs , as well as some disparaging of those who invest time and money in pet dogs . At the organizational level , participants found the campaign information to be insufficient and ill-timed , and campaign locations and personnel inadequate . At the community level , the influence of landscape and topography on accessibility to vaccination points was reported differently between participants from the urban and peri-urban areas . Poor security and impermanent housing materials in the peri-urban areas also drives higher prevalence of guard dog ownership for home protection; these dogs usually roam freely on the streets and are more difficult to handle and bring to the vaccination points . A well-designed communication campaign could improve knowledge about canine rabies . Timely messages on where and when vaccination is occurring could increase dog owners’ perception of their own ability to bring their dogs to the vaccination points and be part of the campaign . Small changes in the implementation of the campaign at the vaccination points could increase the public’s trust and motivation . Location of vaccination points should take into account landscape and community concerns .
The city of Arequipa , Peru is in the midst of an urban rabies epidemic . The first rabid dog was detected in March 2015 , a rare instance of canine rabies reintroduction into an area previously declared free of transmission [1] . To date no human cases have been detected in Arequipa; however , continued transmission in dog populations puts the almost one million inhabitants of the city at risk of infection . Annual mass dog vaccination campaigns were instrumental in eliminating the disease from Arequipa in the 1990s . Unfortunately , this achievement was followed by low vaccination coverage in the years preceding the current outbreak [2] . Following the reemergence of the rabies virus in the city of Arequipa , the Ministry of Health of Peru ( MOH ) initiated additional vaccination campaigns with varying frequency [3] . These efforts have failed to quell the epidemic in Arequipa’s complex urban environment . Particularly , the city’s landscape is characterized by the presence of large dry water channels crossing the city in which free-roaming owned and stray dogs move , mix , and breed . Mass dog vaccination remains the most effective strategy to eliminate canine rabies and canine-mediated human rabies in developing countries [4–6] . In Latin America , a combination of intensive dog vaccination and surveillance efforts has produced dramatic decreases in canine and human cases [5] . However , in addition to reintroduction in Arequipa , canine rabies has been recently introduced to new areas in Latin America such as Jujuy and Salta in Argentina , Mato Grosso do Sul in Brazil and Loma Plata in Paraguay [1] . New efforts and strategies to control the disease in the region are critical . Eliminating canine rabies is feasible; however , several attempts based on dog vaccination have failed to achieve adequate coverage in the Americas , Africa and Asia [5 , 7 , 8] . There are multiple potential drivers of low participation in vaccination campaigns in southern Peru and elsewhere . The social ecological model [9] is a theory-based model that recognizes the complexity of the socio-cultural system in which individuals make decisions and take actions . This theory emphasizes that individuals not only make decisions based on their own knowledge and experience ( individual level factors ) , but are also influenced by their interpersonal relationships ( e . g . norms , families and peers ) , organizations ( e . g . health promotion and prevention activities of health services ) , their community ( e . g . physical environment ) , and policies ( e . g . national or state laws ) [9] . At the individual level , the ability to restrain and handle dogs [10–15]; lack of time to attend vaccination campaigns [12 , 13 , 15]; lack of information [10 , 13–15]; and level of knowledge of rabies [14 , 16–20] have been shown to influence vaccination uptake . At the interpersonal level , social norms ( i . e . those affecting dog ownership practices ) and migration patterns can have a distinct impact on people seeking preventive services , such as vaccines for their dogs [21–25] . At the organizational level , location and number of rabies vaccination points may be particularly important [11 , 15 , 16] . Also , the quality and quantity of health messages about rabies and the rabies campaigns may impact people’s knowledge , but not necessarily their behaviors [26] . At the community level , distance and topography can act as barriers to providing and accessing health services in urban and peri-urban settings [27–30]; in rural settings in particular , distance has been reported as a barrier to achieving high dog rabies vaccination coverage [11 , 15 , 16 , 31] . At the policy level , efforts to eliminate canine rabies have been jeopardized by lack of funding and low political will [5 , 32] . In Peru , the MOH organizes annual mass canine rabies vaccination campaigns in all cities ( affected and unaffected by canine rabies ) . Campaigns are held on either one or two weekend days , offer free vaccination in outdoor settings , and are voluntary . The MOH uses a cell-culture-based vaccine [33] . Campaign promotion is conducted at the local level , primarily via posters placed in health centers , corner stores , schools and other meeting places a couple of weeks or a few days before mass vaccination . The day before and the same day of mass vaccination , campaign staff promote the campaign on foot or from trucks using megaphones; their routes are somewhat haphazard . Locations of the vaccination points are determined a few days in advance; some locations are relatively permanent from year to year ( e . g . the entrance to a health post ) , whereas others may be selected on the day of the campaign . Teams may also move during the course of the day from a lower-demand to a higher-demand location . Vaccination usually starts between 8:30 and 10:00 am , and ends between 1:00 and 3:00 pm . The dog population ( coverage denominator ) is estimated based on the human:dog ratio method . The true human:dog ratio is highly variable geographically [34] and the ratio used throughout Arequipa has been changing from 10:1 to 5:1 in the last two years making vaccination coverage estimation a fast moving target . The reemergence of rabies in the city of Arequipa has been associated with low dog vaccination coverage [2] and a high density of free-roaming dogs ( i . e . stray and owned dogs that spend unsupervised time in the streets and water channels ) [2 , 3] . The social and physical aspects of urbanization in rapidly growing cities such as Arequipa may also facilitate the emergence of canine rabies and complicate its control [35–39] . Epidemics of rabies and other zoonotic pathogens are ongoing in major urban centers across Latin America and worldwide [5 , 40–46] , and it is necessary to understand barriers to dog vaccination , as well as to assess people’s understanding of rabies transmission and prevention . The objective of this study was to qualitatively assess barriers to and facilitators of dog vaccination during the mass campaigns implemented by the MOH in Arequipa , Peru , as well as to explore strategies to increase participation in future campaigns .
Institutional Review Board approval was obtained from Universidad Peruana Cayetano Heredia ( approval identification number: 65369 ) , Tulane University ( approval identification number: 14–606720 ) , and University of Pennsylvania ( approval identification number: 823736 ) . The study was conducted in the Mariano Melgar district ( pop . 55 , 000 ) of the city of Arequipa , Peru’s second largest city . Arequipa is home to 969 , 000 people [47] , and is situated at ~2 , 300 meters above sea level . The first detection of a rabid dog in the city of Arequipa occurred in March 2015 and by January 2016 ( when our data were collected ) , 20 rabid dogs had been detected , of which 11 were found in Mariano Melgar . The city of Arequipa comprises communities spanning different stages of urbanization and different migration histories , from old established neighborhoods , to young neighborhoods , to recent invasions [48] . Within this gradient of development , young neighborhoods and recent invasions are often located on the periphery of the city ( peri-urban area ) and the older localities are nearer to the center ( urban area ) [38] . Compared to the urban area , peri-urban areas generally have lower socioeconomic status , fewer community resources , more security problems , and often more rugged and uneven terrain ( Fig 1 ) . As new neighborhoods mature into established neighborhoods with wealthier residents , homes are improved with better quality construction material and permanent utility connections , and connectivity with the rest of the city increases with better sidewalks , roads , and transportation access . One of Arequipa’s 14 districts , Mariano Melgar transects the city , running from the center to the periphery; the district continues to grow towards the outskirts of the city . In our study , participants represented either the urban or peri-urban areas of the city of Arequipa . The urban neighborhoods were founded several decades ago , while the peri-urban neighborhoods in our study originated around 2000 or later . Purposive sampling was used to select participants for a series of eight focus groups ( FG ) . All participants were residents of urban and peri-urban neighborhoods and were recruited according to their geographic proximity to the site where the focus groups were held . To ensure that participants were not overly exposed to health promotion messages ( i . e . posters about rabies vaccination campaign ) or activities from a nearby health facility , all participants were recruited from homes located at least 6 blocks from the district's municipal building and health posts . The population in the Mariano Melgar district that lives 6 blocks or further from a health post or the municipal building was approximately 45 , 256 people . All participants in each focus group lived in the same neighborhood , but were recruited to represent a range of ages , gender , dog ownership ( i . e . most participants had dogs , but in each FG we tried to have one person who did not currently own a dog ) , and dog vaccination status ( i . e . we aimed to include both dog owners who had and had not participated in the vaccination campaigns in each FG ) . Research team members recruited participants door-to-door one day prior to the focus group . Pre-selecting blocks from a map , research assistants visited every third house within a block until they identified 2 or 3 individuals 18 years or older , preferably a dog owner , who were willing to participate . The same strategy was applied to all selected blocks until 15 individuals were invited to each focus group , to ensure a target attendance of 8–12 participants . The recruitment team explained the study goals and obtained written informed consent from all eligible participants . The recruitment team made follow-up visits the next day 45 minutes before the agreed-upon focus group time to remind participants . Consenting participants were picked up by car from their homes 10 to 15 minutes before the beginning of the FG and taken to the FG location . Participants were compensated for transportation home . Seventy individuals ( 18 to 81 years of age ) participated in eight FG ( 4 groups each in the urban and peri-urban communities ) over four days . The majority of participants were female ( n = 54 ) , but all focus groups had female and male participants ( Table 1 ) . FG guides were developed to cover four topics: dog ownership , dog ecology , and barriers to and facilitators of dog vaccination . Four facilitators conducted the FG: a social scientist experienced in qualitative research ( VPS , PhD in public health ) , a research assistant ( JB , BA in psychology ) , an infectious disease scientist ( RCN , veterinarian and PhD in epidemiology , lead investigator of the study ) , and a note taker . All facilitators were Peruvian , and the infectious disease scientist has lived and/or worked in Arequipa for more than seven years . All FG discussions were digitally audio-recorded and transcribed , and detailed notes were taken throughout . An inductive coding process was used: we first grounded ourselves in the data to explore the topics that would emerge , not knowing what to expect [49] . Codes were then developed based on the emerging themes . Data were imported into ATLAS . ti [50] and coded by two members of the research team; any inconsistencies in the coding were discussed thoroughly until agreement , with codes added as needed . Transcripts that had been coded were recoded with the revised coding scheme . To address the main question of interest to the researchers , all data coded to themes related to canine rabies vaccination barriers and facilitators were summarized , stratified by urban vs . peri-urban communities , and younger ( <30 years of age ) and older participants .
Various theories have been developed to guide assessments of effectiveness of rabies vaccination campaigns [10 , 52 , 53]; an important component of this work is examining the perspectives of dog owners and community members , both qualitatively and quantitatively [53] . This article analyzes unique focus group data , and is the first from our ongoing multi-method studies of factors associated with coverage of and participation in mass canine rabies vaccination campaigns . The social ecological model [9] applied here to structure the themes raised by FG participants is a theory-based model that recognizes the complexity of the socio-cultural system in which individuals make decisions and take actions . Our findings reveal multiple barriers to dog vaccination at the individual , interpersonal , organizational and community levels of the social ecological model . We identified individual-level barriers to vaccination as well as facilitators of vaccination . An important next step is to translate these barriers and facilitators into proposed solutions to improve campaign effectiveness , using evidence-based concepts from health behavior change theory . For example , our respondents suggested using fear-based promotional materials and punishments for non-participation . This proposed solution highlights the importance of self-efficacy ( the self-perceived ability that one can practice a proposed solution ) as a driver of health behavior , as posited in several health behavior change theories , including the extended parallel processing model ( EPPM ) [54] . The EPPM model proposes that when faced with a health risk , people will take actions to protect their health if they are empowered by an appropriate balance of fear and self-efficacy [54] , consistent with other models that predict adverse consequences from employing fear in health communication programs without assuring high self-efficacy levels [55–58] . A health communication campaign that focuses on instilling fear alone will often result in a feeling of helplessness and inaction . However , communication that conveys risk in an understandable way and also provides a clear and do-able call to action to reduce one’s risk leads to empowered citizens who take action [59] . Rabies is a fatal and devastating disease and Arequipa is facing a sustained rabies outbreak that has not yet been controlled , posing a serious risk for humans and animals; vaccinating one’s dog is a highly effective prevention method that only requires dog owners’ efforts once a year . Rabies vaccination communication should tap into this feeling of risk ( to increase individuals’ perceived susceptibility to rabies , and inform them of the severity of this virus ) , while at the same time help individuals feel confident in being able to act ( self-efficacy in handling their dogs and participating in campaigns ) to effectively mitigate that risk . This communication should not only address knowledge gaps , but also inspire trust in the program and emphasize motivators that participants described , including keeping one’s family healthy , protecting the household pets from rabies , and keeping the community free of rabies . A very salient individual level barrier mentioned in FG was the logistics: difficulty getting dogs to vaccination sites , dog fights en route to and at the vaccination points . To a certain extent , this barrier can also be mitigated through improvements to the rabies vaccination program . Other Latin American countries have taken steps to improve people’s ability to handle their dogs . For instance , Costa Rica included in their “Guidelines of Pet Reproduction and Ownership” to walk dogs on the leash frequently to increase their proper socialization [60] , and new Mexican bills support citizens’ initiatives to adequately integrate dogs in society , such as the “Social Dog , Responsible Owner Program” [61] , which provides training to dog owners and their dogs through in-person workshops , online videos , and written material . Based on our findings , to address some of the knowledge gaps in the population that could deter vaccination for some dogs , health communication campaigns should include additional information about when one should vaccinate one’s dogs and describing that all dogs ( even those that are always indoors ) are at risk . At the interpersonal level , barriers discussed by participants were related to the social norms of poor animal care culture in Arequipa , competing social pressure both for and against vaccination , although the latter was considerably less common . Most respondents in this study also indicated that there is social pressure in Arequipa for people to vaccinate dogs , as those who do not vaccinate are seen in a negative light; they are considered lazy , uncaring for animals , and irresponsible neighbors in the context of community protection offered by vaccines ( although a few also described social pressure to not vaccinate—feeling laughed at for spending time and/or money on a dog ) . The theories of diffusion of innovations [62] , social learning [63] and planned behavior and reasoned action [64] point to the importance of social norms , the opinion of others , and social networks in making decisions about health behavior . Vaccine promotion communication in Arequipa should be informed by these theories and seek to influence local norms by providing information from trusted and influential sources , as well as identifying well-connected individuals in communities to champion certain ideas—including but not limited to local storekeepers , health workers that routinely work in certain areas , health care providers ( i . e . veterinarians ) , local neighborhood authorities , municipal governments , health officials , local celebrities , teachers , and respected elders . Barriers that emerged at the organizational level were health system weaknesses , such as insufficient promotion of the vaccination campaign , poor selection of location for the vaccination points , low frequency of vaccination campaigns , and insufficient staff during the campaign . Regarding the service , trust in quality of services can be improved through strategic decisions and actions in the implementation phase . Identification cards with the vaccinator’s name and qualifications should be visible to campaign attendees—both on the vaccinator and on any visible media ( e . g . posters , banners ) at the site . Likewise , keeping coolers visible and less exposed to the sun might help increase confidence that the cold chain has been kept intact , showing the unopened syringe or needles before filling them would quell rumors of syringes are reused , and pointing out the expiration date on the vaccine vials before filling the syringe would take seconds , but increase trust in the product . Finally , a barrier mentioned in many FG was the insufficient or untimely information about date and locations of vaccine campaigns . This barrier can also be addressed in the planning and implementing phases: announcements about the campaign can start earlier and through different media , and be conducted on different weekdays and times , to ensure more people hear about them . Posting times , dates , and locations of upcoming vaccination campaigns in advance and in visible community locations would also allow for individuals to plan where and when to take their dogs for the vaccination , and to plan to have extra people to help take the dogs if necessary . Participants mentioned other health programs based on door-to-door interventions such as the Chagas control campaign that could have lessons to offer the rabies campaign . Door-to-door strategies have been discussed among local officials to promote participation and also enable a simultaneous canine census to improve the accuracy of coverage data . These and other interventions at the organizational level to improve rabies vaccination uptake can be modeled according to the health care access framework presented by Obrist et al . [52] . This framework consists of five dimensions: availability , accessibility , affordability , adequacy , and acceptability; the present study provides insights into each of these dimensions . At the community level , barriers cited included distance to vaccination points , difficult topography ( particularly in the peri-urban areas ) , and lack of security in the area . Flexible strategies are necessary to meet the local community needs [11 , 16]; for instance , in rural Tanzania [11] , a few central-point canine rabies vaccination points were insufficient to achieve 70% coverage ( WHO recommended threshold ) , and had to be supplemented by door-to-door efforts [65] . The combination of central-point and door-to-door dog vaccination has also proven to be effective in urban environments , with one large city recently achieving 79% coverage [66] . Participants’ preference for door-to-door campaigns is also important to highlight: this has usually been done in rural areas [11 , 66] , and although the topography and urban design in some cities is ideal for centralized vaccinations , clearly not all urbanized settings are the same: a central-point mass vaccination campaign in the urban areas of N’Djamena , Chad achieved a coverage above 70% [67] , while a similar central-point vaccination campaign in the city of Bamako , Mali achieved only 17% coverage [10] . In our study , campaigns in the urban areas were fairly accessible to dog owners ( despite having to cross large avenues ) , but the topography of the peri-urban areas was much more difficult ( steep hills , large rocks , unpaved roads ) for people trying to get multiple unleashed dogs to vaccination points . Rapid urban spread , low salaries , scarce resources and bureaucratic rules can all impede the responsiveness and quality of campaign services [68] . For rabies , control programs must take into account social , political and cultural contexts to improve efficacy and avoid the barriers faced by other top-down public health interventions [69–71] . Our study and others [16–18 , 26 , 72–74] have focused on dog owners’ knowledge , dog-ownership practices , norms , and perceptions about the dog vaccination campaign as important factors to understand the rabies program’s outcomes . But capacities , norms , and policies of implementing institutions also play a central role in rabies control efforts [16] . Overall , although some of the suggestions discussed would add to program expenses , several imply little marginal cost; many suggested strategies simply optimize current investments for better efficiency . For example , current promotional materials could be more informative ( i . e . include location and hours of vaccinations ) and be disseminated earlier; vaccinators could visibly display their ID and credentials; and vaccine syringes could be opened in front of dos owners with additional explanations about safety and quality procedures to increase trust in the campaign . Similar to prior studies [74] , we found different attitudes , dog-ownership practices , and knowledge levels between urban and peri-urban areas . Given the different characteristics of these areas ( topography , migration history , level of urbanization , levels of security and utilization of dogs for companionship vs . protection ) , it is difficult to justify a one-size-fits-all campaign strategy . The newer peripheral communities we studied in Arequipa are typical of peri-urban areas , a neglected zone in Latin America , Africa , and Asia [75] commonly characterized by inadequate infrastructure , service provision , and land tenure security [38 , 76] . While most public health campaigns have distinct strategies for urban vs . rural areas; the peri-urban area ( or rural-urban interface ) , one of the fastest growing areas in the world [75] , lacks a specific and responsive programmatic approach in Peru and elsewhere .
Important individual , interpersonal , organizational , and community level factors limit more widespread participation in dog rabies vaccination campaigns in the city of Arequipa–site of a canine rabies outbreak since 2015 . A comprehensive communication campaign is required to increase the population’s knowledge about rabies’ transmission , its consequences , and prevention measures . Developing strategies to help dog owners not accustomed to using leashes should be explored and implemented to help facilitate the transport of dogs to vaccination points . It is important to provide timely information about free dog mass vaccination campaigns . Clear identification of vaccinators and demonstration of unopened needles and syringes , as well as training vaccinators on their interactions with dog owners and response to their concerns can also help increase trust in the campaign . Finally , flexible strategies are needed to serve diverse communities within a city; urban and peri-urban areas present contrasting landscapes that might require different vaccine point locations . | Canine rabies was reintroduced in Arequipa , Peru in March 2015 , a rare event in an area previously declared free of transmission . In Arequipa , annual mass dog vaccination is practiced as a preventive strategy , with additional campaigns being implemented since the recent detection of the virus . However , these additional efforts have not quelled the outbreak and low dog vaccination coverage is driving ongoing transmission . We conducted focus groups in urban and peri-urban areas of Arequipa to identify barriers to and facilitators of canine vaccination during mass campaigns . Based on our findings , communication campaigns should seek to increase knowledge about canine rabies and the vaccination campaign , and provide timely messages on where and when vaccination is occurring . Small changes at the campaign’s vaccination points could increase public’s trust . Finally , there are differences between urban and peri-urban areas , such as landscape and topography that affect participation in mass vaccination campaigns and that should be considered when selecting locations for vaccination points . | [
"Abstract",
"Introduction",
"Methods",
"Discussion",
"Conclusions"
] | [
"animal",
"types",
"medicine",
"and",
"health",
"sciences",
"ecology",
"and",
"environmental",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"immunology",
"tropical",
"diseases",
"microbiology",
"vertebrates",
"pets",
"and",
"companion",
"animals",
"dogs",
"animals",
"mammals",
"vaccines",
"preventive",
"medicine",
"viruses",
"rabies",
"rna",
"viruses",
"neglected",
"tropical",
"diseases",
"infectious",
"disease",
"control",
"vaccination",
"and",
"immunization",
"zoology",
"rabies",
"virus",
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"geography",
"zoonoses",
"medical",
"microbiology",
"microbial",
"pathogens",
"community",
"ecology",
"lyssavirus",
"urban",
"areas",
"ecology",
"earth",
"sciences",
"geographic",
"areas",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases",
"amniotes",
"organisms"
] | 2017 | Barriers to dog rabies vaccination during an urban rabies outbreak: Qualitative findings from Arequipa, Peru |
Cyanide-resistant non-phosphorylating respiration is known in mitochondria from plants , fungi , and microorganisms but is absent in mammals . It results from the activity of an alternative oxidase ( AOX ) that conveys electrons directly from the respiratory chain ( RC ) ubiquinol pool to oxygen . AOX thus provides a bypath that releases constraints on the cytochrome pathway and prevents the over-reduction of the ubiquinone pool , a major source of superoxide . RC dysfunctions and deleterious superoxide overproduction are recurrent themes in human pathologies , ranging from neurodegenerative diseases to cancer , and may be instrumental in ageing . Thus , preventing RC blockade and excess superoxide production by means of AOX should be of considerable interest . However , because of its energy-dissipating properties , AOX might produce deleterious effects of its own in mammals . Here we show that AOX can be safely expressed in the mouse ( MitAOX ) , with major physiological parameters being unaffected . It neither disrupted the activity of other RC components nor decreased oxidative phosphorylation in isolated mitochondria . It conferred cyanide-resistance to mitochondrial substrate oxidation and decreased reactive oxygen species ( ROS ) production upon RC blockade . Accordingly , AOX expression was able to support cyanide-resistant respiration by intact organs and to afford prolonged protection against a lethal concentration of gaseous cyanide in whole animals . Taken together , these results indicate that AOX expression in the mouse is innocuous and permits to overcome a RC blockade , while reducing associated oxidative insult . Therefore , the MitAOX mice represent a valuable tool in order to investigate the ability of AOX to counteract the panoply of mitochondrial-inherited diseases originating from oxidative phosphorylation defects .
In mammalian mitochondria , the terminal step of electron transfer to molecular oxygen , producing water , is exclusively mediated by the cyanide-sensitive cytochrome c oxidase ( COX ) [1] and the electron transfer is tightly coupled to proton translocation . Protons simultaneously accumulated on the outer surface of the inner membrane are subsequently used by the ATP synthase ( complex V , CV ) to generate ATP from ADP and inorganic phosphate imported in the mitochondrial matrix by the adenylate carrier ( Ant ) and the phosphate carrier ( Pic ) respectively [2] ( Figure 1A ) . Usually , a small percentage of electrons escapes from the RC to produce superoxide , with proposed roles in metabolic signaling [3] . However , conditions leading to the over-reduction of the ubiquinone pool may result in the production of excess superoxide , with deleterious consequences [4] , [5] . In plants , many microorganisms , and a few animals [6] , a non proton-motive , cyanide-resistant AOX , can also oxidize ubiquinol to produce water [7] ( Figure 1A ) , maintaining electron transfer even when the activity of the cytochrome segment of the respiratory chain ( namely complex III to IV ) is limiting or unavailable [8] . Under such conditions , AOX also prevents the over-reduction of ubiquinone , serving , in effect , an antioxidant role [9] . Crucially , the enzymatic properties of AOX ( low ubiquinol affinity ) tend to limit its involvement in respiration in vivo to conditions of substantial over-reduction of the quinone pool , minimizing detrimental competition with the phosphorylating cytochrome pathway [10] . Nevertheless , in case of blockade of the cytochrome pathway , AOX enables divalent electron flow to oxygen , thus acting as a safety valve to preserve respiration , restore metabolic balance , and minimize excessive superoxide production [11] . Accordingly , we previously showed that Ciona intestinalis AOX could be expressed in cultured human cells , conferring cyanide-resistant respiration without harmful effects [12] and counteracting the consequences of genetic defects in COX [13] . Similarly , viable and active flies ubiquitously expressing AOX and substantially resistant to the action of antimycin ( a complex III-specific inhibitor; Figure 1A ) or cyanide were obtained [14] . AOX expression in flies also rescued the lethality of genetically-induced COX deficiency [14] . Altogether , these findings were an incentive to attempt AOX expression in vivo in a mammal . Here we show that the AOX can be expressed safely in a mammal , without any obvious detrimental effect on major physiological parameters . We show also that the presence of the AOX conferred cyanide-resistance to mitochondrial substrate oxidation and decreased ROS production under conditions of RC inhibition . Importantly , the AOX also conferred cyanide-resistance to intact organ respiration and significantly prolonged the survival of the whole organism in the presence of this deadly poison . Taken together , these results indicate that the AOX is active in vivo in the MitAOX mouse and counteracts respiratory chain blockade and its physiological consequences .
The Ciona intestinalis AOX gene was recoded to maximize its expression in the mouse and introduced into early mouse embryos by germ-line lentiviral transduction on a mixed genetic background ( CD-1/B6 ) . We used the ubiquitously active , chimeric CAG promoter , together with the Woodchuck hepatitis virus Post-transcriptional Regulatory Element ( WPRE ) to further enhance AOX gene expression ( Figure 1B ) . PCR analysis of genomic DNA up to the F3 generation indicated the presence of the AOX transgene in all founder descendants . Copy number was estimated by Southern blot at 4 to 8 per genome in the founders ( Figure 1C ) . Western blot analysis of the F3 generation brain mitochondria indicated a consistent level of AOX protein between siblings ( Figure 1D ) . Litter size ( 12±2 versus 11±3 in control mice ) was unaffected by the presence of the transgene . A number of F3 individuals were analyzed for AOX distribution pattern . Western blots indicated , similarly to F1 individual ( Figure 2A ) , widespread tissue expression , with expression prominent in brain and pancreas and varying among different tissues ( Figure 1E ) . We also checked the stability of AOX expression as a function of age and observed a preserved AOX expression in all the different tissues studied in 15 month-old animals ( Figure 2A ) . We next showed that the presence of the AOX did not alter the steady state levels of the different RC complexes ( Figure 2A–2B ) . Noticeably , AOX did not require tight association with any of the RC complexes or supercomplexes in order to be functional , since the enzyme was not found associated with these entities in BN-PAGE analyses ( Figure 2C ) . AOX migrated as a dimer with an apparent molecular weight of 70–72 kDa , with a substantial proportion found as higher polymeric forms , mostly tetramer , or tending to aggregate under this condition ( 6 g/g digitonin ) , as previously observed in organisms where AOX is naturally present [15] . Finally , we showed that neither the distribution nor the quantities of the RC supercomplexes were significantly modified by the presence of the AOX ( Figure 2D ) . Detailed immunohistological analysis of highly AOX-expressing tissues revealed differential expression depending on organ sub-territories . For instance , in the pancreas , the pattern of AOX expression matched that of complex III Core protein I ( Core I ) and COX I ( complex IV; not shown ) , being much higher in exocrine than in endocrine tissue ( insulin-producing Langerhans islets; Figure 3A ) . In the brain ( Figure 3B ) , AOX was massively expressed in the CA3 pyramidal layer and the cortex , with a perfect overlap with COX I ( or Core I , or ATPase α ( complex V ) ; not shown ) antibody staining . AOX expression was lower in the lateral amygdalar nucleus , even lower in the CA1 pyramidal layer , and hardly detectable in the thalamus , the hypothalamus , or the granule cell layer of the dentate gyrus , despite strong staining with COX I ( Figure 3B ) or other OXPHOS marker antibodies ( Core I , ATPase α; not shown ) . Interestingly , immunohistological study performed on the brain of WT and MitAOX mice using COX I antibody revealed similar expression , which denotes the absence of detectable effect of AOX expression on the amount and distribution of mitochondria in the brain ( Figure S1 ) . Mitochondria were next isolated from a number of tissues as to investigate AOX functionality . A significant cyanide-resistant oxidation of succinate was detected in several tissues and was proportional to the AOX protein level . For instance , tissues with the highest protein expression ( brain and pancreas ) , showed the highest cyanide resistance , which was less in heart and nearly inexistent in liver ( Figure 4A , traces a , c , e , g , h ) with the lowest protein level . In all cases , cyanide resistant respiration was fully inhibited by 50 µM propylgallate ( PG ) , a specific inhibitor of AOX . A quite similar cyanide-resistance ( about 30% ) was measured using malate plus glutamate as substrate ( Figure 4A , trace c ) . Noticeably , the oxidation of malate plus glutamate was still efficiently controlled by the phosphorylation process in the MitAOX mouse . This is shown by the large stimulation of malate oxidation triggered by the ADP addition in the presence of cyanide ( Figure 4A , trace d ) . Because AOX expression in the brain or in the pancreas ( Figure 3 ) was not uniform , the true extent of cyanide resistance in AOX-expressing sub-territories in these tissues is presumably even higher . In comparison , whatever the substrate being oxidized ( succinate or malate ) by mitochondria of WT animal tissues , cyanide fully inhibited oxygen consumption ( less <1% resistance to cyanide; Table S1 ) . Afterwards , using a homemade device supporting a nylon net , we were able to study the whole organ respiration using the Clark oxygen electrode chamber . As compared to WT , a significant cyanide-resistance of whole organ respiration , 30% and 50% for brain hemisphere and optic nerve respectively ( Figure 4B , traces a–d ) was observed in MitAOX . In order to estimate indirectly the participation of AOX in the oxidation of succinate under phosphorylating conditions ( presence of ADP ) we determined the ADP/O values in WT and MitAOX mice brain mitochondria ( Figure 5A ) . Any significant involvement of the non-proton motive AOX in electron flow should decrease the use of ADP associated with O2 consumption ( Figure 1A ) , thereby diminishing the ADP/O ratio . The measured values ( ≈1 . 4; Figure 5A ) were found to be similar in MitAOX and WT brain mitochondria , indicative of a negligible participation of AOX in electron transfer in the presence of ADP , as previously reported for mitochondria of organisms naturally endowed with AOX [16] . Extensive investigations of brain and pancreas mitochondria revealed no significant impact of AOX expression on RC activities ( Figure S2A ) . Because the low O2 level in vivo ( i . e . 25 to 40 µM in the brain ) might affect the activity of the AOX , we next tested cyanide-resistance of substrate oxidation as a function of O2 concentration ( Figure 5B ) . We observed a gradual decrease of cyanide-resistance only for O2 concentrations below 20 µM . At these low oxygen tension values , percent of cyanide-resistance was confirmed by the simultaneous measurement of oxygen consumption in a closed chamber by a classical Clark electrode and a fluorescence-based micro-optode ( Figure S2B , S2C ) . AOX thus appears to be fully functional under physiological conditions . Interestingly , mitochondria from MitAOX mouse brain produced significantly less ROS than their WT counterparts , despite partial expression in the brain ( Figure 3B ) . This was shown by an assay in which superoxide , whose production was triggered by antimycin , was fully converted to hydrogen peroxide ( Figure 5C–5D ) . Noticeably before the addition of antimycin , the limited production of ROS is not affected by the presence of the AOX . In order to evaluate a potential detrimental effect of AOX expression , we next investigated a number of physiological and behavioral variables in pups and mature MitAOX animals ( Table 1 ) . Cardiorespiratory variables displayed minor , albeit statistically significant differences between MitAOX and WT newborn mice . In addition to their slightly smaller weights ( <10% ) , the MitAOX pups had slightly slower heart rates ( <8% ) while minor differences in their breathing pattern did not affect ventilation . Activity and ultrasonic vocalizations ( a common marker of anxiety in newborn rodents ) showed no significant genotype-related differences . Thus , the analysis of physiological and behavioral variables in MitAOX newborn mice under normal conditions did not reveal any pathological signs . Later on , at 3 months of age , MitAOX animals had similar weight and performed as WT in the Rotarod test , documenting unaltered motor coordination and fatigue resistance ( Table 1 ) , despite significant AOX expression in the cerebellum and skeletal muscle ( Figure 1E ) . Finally , in order to assess the capacity of the AOX to compensate for a cytochrome pathway blockade in vivo , anesthetized MitAOX mice were exposed to gaseous cyanide . The deadly effect of cyanide on mammals has been previously shown to result mostly from the inhibition of the mitochondrial cytochrome oxidase rather than to its binding to other metalloenzymes [17] . Accordingly , we observed a substantially prolonged survival of MitAOX mice in the presence of a lethal concentration of gaseous cyanide , compared to WT mice ( more than 200% ) . Moreover , by using different transgenic founders with variable AOX expression , we were able to show that the amount of resistance to cyanide in the whole animal is proportional to AOX protein content as determined in the lung and in the brain ( Figure 5E ) .
Our data show for the first time that a functional AOX can be expressed in a mammal and transmitted between generations , conferring significant cyanide-resistance to mitochondrial substrate oxidation and tissue respiration as well as the whole organism . As previously observed in cultured human cells [12] , [18] and flies [14] , [19] , the enzyme is targeted to the mitochondria where it functionally interacts with the RC . Most importantly , our data show that , similarly to the plant enzyme [8] , the C . intestinalis AOX expressed in the MitAOX does not interfere/compete significantly with the cytochrome pathway , being functional only upon blockade of this latter when the pool of ubiquinone becomes highly reduced . Accordingly , the expression of the C . intestinalis AOX in the mouse did not result in any deleterious consequence , whilst spectacularly increasing the survival of the mouse in the presence of a lethal concentration of gaseous cyanide . The protective mechanism provided by the AOX to organisms naturally harboring the enzyme was therefore fully preserved when the oxidase was expressed in the mouse . The AOX protein typically has such a high Km for reduced quinones ( apparent KmDQH2 from 0 . 53 to 0 . 38 mM ) [20] that it competes only very poorly with the cytochrome pathway ( apparent KmDQH2 less than 20 µM ) [21] for quinol oxidation . In plants , such a competition is further avoided by a specific channeling of electrons , dictated by the structural association of the relevant electron carriers , involving the malic enzyme , the Ndi ( internal rotenone-insensitive NADH dehydrogenase ) , and the AOX proteins [22] , [23] . Noticeably , in the absence of Ndi in the MitAOX mouse , the non proton-motive AOX can still promote ATP formation through activation of NADH oxidation by the proton-motive complex I . The control of superoxide overproduction by AOX [24] , [25] illustrates a second protective effect resulting from the AOX expression in the mouse . According to our data , C intestinalis AOX efficiently decreases superoxide overproduction triggered by the over-reduction of the ubiquinone pool , which was generated by antimycin . On the other hand , the innocuousness of AOX expression in the mouse suggests that any production of superoxide that is physiologically required [26] , [27] is not significantly modified by the presence of a functional AOX in the mitochondria . The observed protection provided by the AOX against the toxic effects of cyanide or antimycin , in the light of its expression territories , should enable the use of the MitAOX mouse to testing the potentially deleterious role of mitochondrial dysfunction and/or the resulting oxidative stress in mouse models of neurodegenerative diseases [28] . It will be similarly interesting to use the MitAOX mouse to investigate the potential role of mitochondrial dysfunction and excess ROS production in ageing [29] . Indeed , a single polypeptide , AOX , can replace two elaborate multisubunit complexes ( complexes III and IV; 11 and 13 subunits respectively ) , without competing with these under normal conditions . AOX expression may thus also chart a way to implementing a wide-spectrum therapy for currently intractable but major disease entities [11] . Finally , we may wonder why the AOX , with the huge metabolic flexibility that it confers to the cell , has been lost in most of the animal kingdom . One intriguing clue comes from the fact that organisms naturally endowed with AOX are almost exclusively sessile or pelagic , and their mitochondria must regularly endure harsh energetic and stress conditions from which they cannot escape: e . g . , activation of photosynthesis for plants [30] , [31] , exposure to toxic xenobiotics for microorganisms , or local fluctuations in the marine environment ( temperature , oxygen , nutrient levels ) for animals that are fixed in one place [6] . In contrast , in fast-moving organisms , AOX activity would only be advantageous under peculiar conditions , such as those arising in cases of mitochondrial diseases where OXPHOS is primarily or secondarily affected . The creation of a mammal expressing the AOX will surely be crucial in shedding light on this puzzling evolutionary interrogation .
The mice were housed with a 12-h light/dark cycle and free access to food ( 3% lipids , 16% protein; SAFE A-04 chow; UAR Epinay sur Orge , France ) and water . Animal management was in accordance with Good Laboratory Practice Guidelines [32] . All experiments were carried out following the recommendation of INSERM for the use of animal laboratory and the approval by the ethical committee of Debre-Bichat Hospitals; project number 2010-13/676-0014 . The Ciona intestinalis AOX cDNA sequence [12] was redesigned , optimized ( AOXopt ) in order to ameliorate its expression in mice ( DNA2 . 0 algorithm ) and synthesized by DNA2 . 0 ( Menlo Park , CA , USA ) . The redesigned cDNA was flanked by the attL1/L2 recombination sites . Next , using Gateway cloning technology , the AOXopt cDNA was transferred by an in vitro one-step recombination to the attR1/R2-containing plasmid pTrip [33] , further used to produce the AOX lentiviral vectors . Afterwards , the strong chimeric CAG promoter was isolated as a 1 . 7 kb digestion fragment ( EcorV-Mlu1 ) from the p97 Vector and cloned upstream the AOX in the pTrip-AOXopt vector , giving rise to pTrip-CAG-AOXopt . Finally , the Woodchuck Hepatitis post-transcriptional Regulatory Element ( WPRE ) was isolated as a 650 bp digestion fragment ( BstX1-Kpn1 ) from plasmid pT45 and introduced downstream of the AOXopt cDNA , giving rise to the pTrip-CAG-AOXopt-WPRE plasmid . Lentiviral vectors containing the CAG-AOXopt-WPRE construct were generated as previously described [33] . Before injection , the HIV p24 Gag antigen was quantified by ELISA ( HIV-1 P24 antigen assay; ZeptoMetrix corporation , NY , USA ) , and the AOX-expressing lentiviral vectors were titered by transducing 40 , 000 HeLa cells in 24-well plates with serial dilution ( 2 , 1 , 0 . 5 µl ) . Each early mouse embryo was injected with 50–500 pL ( p24 vector titer: 109 ng/µl ) . Mouse genomic DNA was extracted from frozen tail samples using the MasterPure DNA Purification Kit according to the manufacturer's instructions ( Tebu-Bio , Le Perray en Yveline , France ) . The transmission of the AOX transgene was verified by PCR on tail genomic DNA using AOXopt-F ( GGATGAGCCCAATATCGAAG ) and AOXopt-R ( CTGAAACGAAAATGCCTTGG ) primers . For Southern blot analysis the DIG System from Roche Applied Science was used . Western blot analyses were performed as indicated in [13] . In addition , blots were re-probed with an anti-ßATPase ( 1∶5000 , rabbit polyclonal antibody raised against the yeast ßATPase and kindly provided by A . Tzagoloff ) as a standardization control . Peroxidase-conjugated anti-rabbit secondary antibody ( 1∶5 , 000 , Amersham , Buckinghamshire , UK ) was used at 5 , 000-fold dilution . Blue-native PAGE ( BN-PAGE ) analyses were performed on isolated mitochondria as described [34] . Concentration of detergents as indicated in the figure legend . Respiratory chain enzyme activities were spectrophotometrically measured using a Cary 50 UV–visible spectrophotometer ( Varian Inc , Les Ulis , France ) [35] . Mitochondrial substrate oxidation was polarographically estimated using a Clark oxygen electrode ( Hansatech Instruments , Norfolk , England ) in a magnetically-stirred chamber maintained at 37°C in 250 µl of a respiratory medium consisting of 0 . 3 M mannitol , 5 mM KCl , 5 mM MgCl2 , 10 mM phosphate buffer ( pH 7 . 2 ) and 1 mg . ml−1 bovine serum albumin , plus substrates or inhibitors as described [36] . Substrate and inhibitor concentrations were as followed: 1 mM ADP , 10 mM succinate , malate/glutamate ( 5 mM each ) , 1 mM KCN , 50 µM PG , 4 µM rotenone . Alternatively , oxygen uptake was measured under similar condition using a micro-optode consisting in an optic fiber equipped with an oxygen-sensitive fluorescent terminal sensor ( FireSting O2; Bionef , Paris , France ) . Superoxide plus hydrogen peroxide production by isolated brain mitochondria was quantified using Amplex Red fluorescence in 1 . 5 ml of medium consisting of 125 mM KCl , 14 mM NaCl , 1 mM MgCl2 , 20 µM EGTA , 4 mM KH2PO4 and 20 mM HEPES ( pH 7 . 2 ) to which were added 2 IU of purified superoxide dismutase [37] . Inhibitory effect of propylgallate could not be tested in these assays because of interactions with the probe . Protein concentration was measured according to the Bradford assay . Breathing variables ( breath duration ( TTOT ) , tidal volume ( VT ) , and ventilation ( VE ) calculated as VT/TTOT ) were measured noninvasively in unanaesthetized , unrestrained 6-day old pups using whole-body flow barometric plethysmography as described previously [40] , [41] . Statistical analyses were performed using Student's t-test ( Statview 5 ) . Values of p<0 . 05 were considered as significant . Motor coordination and fatigue resistance of older animals ( 2 months ) were assessed by Rotarod test as previously described [42] . Cyanide poisoning is classified as an USDA Pain and Distress Category E condition , and the investigators estimated the study acceptable only if the animals were beforehand anesthetized . The investigators realized this might have impacted the outcome of the experiment , but that without the use of anesthesia , the work would have been inhumane . Mice were anesthetized with chloral hydrate ( 350 mg/kg ) . Once anesthetized , one WT and one MitAOX mouse were placed in a 5 . 2 L airtight acrylic glass chamber maintained at 28°C ( above the boiling point of cyanide , 26°C ) . Cyanide gas ( 451 ppm ) was produced in the chamber by injecting 100 mM KCN into a Petri dish containing 10 ml of 1 M sulfuric acid . Respiratory activity of the mice was used as an index of mouse survival . Four experiments were carried out with MitAOX mice endowed with different levels of AOX , afterwards estimated in the lung and the brain by Western blot analysis . | In mammalian mitochondria , the energy-producing machinery is powered by the electron transfer to molecular oxygen , a mechanism whose terminal step is mediated by the cyanide-sensitive cytochrome c oxidase ( COX ) . In plants , fungi , microorganisms , and some lower animals ( but not in mammals ) , in addition to the normal pathway , a cyanide-resistant alternative oxidase ( AOX ) exists . It maintains electron transfer to oxygen even when the normal pathway is blocked . This provides a bypath that releases constraints on the energy producing machinery and prevents the production of deleterious superoxide molecules . Thus , preventing the energy producing machinery blockade and excess superoxide production by means of AOX should be of considerable interest . However , because of its energy-dissipating properties , AOX might produce deleterious effects of its own in mammals . Here we show that the AOX can be safely expressed in a mammal with major physiological and molecular parameters being unaffected . We also show that the AOX is active in vivo where it counteracts the energy producing machinery blockade and reduces in vitro the associated oxidative insult . Up to now , efficient therapies against mitochondrial-associated diseases are lacking dramatically . Therefore , in view of our results , the MitAOX mice represent a precious tool to assess the AOX therapeutic capacity against the panoply of inherited mitochondrial diseases . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"animal",
"models",
"biochemistry",
"animal",
"genetics",
"model",
"organisms",
"genetics",
"biology",
"mouse",
"genetics",
"and",
"genomics"
] | 2013 | Alternative Oxidase Expression in the Mouse Enables Bypassing Cytochrome c Oxidase Blockade and Limits Mitochondrial ROS Overproduction |
Two important ideas about associative learning have emerged in recent decades: ( 1 ) Animals are Bayesian learners , tracking their uncertainty about associations; and ( 2 ) animals acquire long-term reward predictions through reinforcement learning . Both of these ideas are normative , in the sense that they are derived from rational design principles . They are also descriptive , capturing a wide range of empirical phenomena that troubled earlier theories . This article describes a unifying framework encompassing Bayesian and reinforcement learning theories of associative learning . Each perspective captures a different aspect of associative learning , and their synthesis offers insight into phenomena that neither perspective can explain on its own .
Learning to predict rewards ( or punishments ) from the occurrence of other stimuli is fundamental to the survival of animals . When such learning occurs , it is commonly assumed that a stimulus-reward association is stored in memory [1 , 2] . Two ideas have , over the last few decades , altered our understanding of how such associations are formed , and the nature of their content . First , Bayesian theories of learning have suggested that animals estimate not only the strength of associations , but also their uncertainty in these estimates [3–8] . Second , reinforcement learning ( RL ) theories have suggested that animals estimate long-term cumulative future reward [9–11] . Both Bayesian and RL theories can be viewed as generalizations of the seminal Rescorla-Wagner model [12] that address some of its limitations . The mathematical derivations of these generalizations and their empirical support will be reviewed in the following sections . Bayesian and RL theories are derived from different—but not mutually exclusive—assumptions about the nature of the learning task . The goal of this paper is to unify these perspectives and explore the implications of this unification . One set of assumptions about the learning task concerns the target of learning . The Bayesian generalization of the Rescorla-Wagner model , embodied in the Kalman filter [3 , 4 , 6] , assumes that this is the problem of predicting immediate reward , whereas RL theories , such as temporal difference ( TD ) learning , assume that the goal of learning is to predict the cumulative future reward . A second set of assumptions concerns the representation of uncertainty . The Kalman filter learns a Bayesian estimator ( the posterior distribution ) of expected immediate reward , whereas TD learns a point estimator ( a single value rather than a distribution ) of expected future reward . As shown below , the Rescorla-Wagner model can be construed as a point estimator of expected immediate reward . After reviewing these different modeling assumptions ( organized in Fig 1 ) , I show how they can be naturally brought together in the form of the Kalman TD model . This model has been previously studied in the RL literature [13] , but has received relatively little attention in neuroscience or psychology ( see [14] for an exception ) . I explain how this model combines the strengths of Bayesian and TD models . I will demonstrate this point using several experimental examples that neither model can account for on its own .
Let xn denote the vector of conditioned stimulus ( CS ) intensities on trial n ( all vectors are taken to be column vectors ) , wn denote the associative strengths ( or weights ) , and rn denote the unconditioned stimulus ( US; i . e . , observed reward ) . Note that traditional associative learning theories interpret rn as the asymptotic level of responding supported by the US on the current trial; however , in this article I interpret rn as reward in order to facilitate the connection to RL . To compactly describe experimental paradigms , I use uppercase letters ( A , B , etc . ) to denote conditioned stimuli , and combinations of letters ( e . g . , AB ) to denote stimulus compounds . A stimulus ( or compound ) terminating in reward is denoted by A→+ . Similarly , a stimulus terminating in no reward is denoted by A→- . A stimulus terminating with the onset of another stimulus is denoted A→B . The notation A→ ? indicates that conditioned responding to A is the dependent measure in a particular experiment . When multiple trial types are interleaved within a phase , forward slashes are used ( e . g . , A→+ / B→- ) , and contiguous phases are separated by semi-colons ( e . g . , A→+; B→- ) . Making predictions about empirical phenomena is complicated by the fact that experimental paradigms use diverse stimuli , rewards , and behavioral measures . The simulations reported below are predicated on the assumption that we can abstract away from some of these experimental details and predict response rates simply on the basis of reward expectation , as acquired by trial-and-error learning . This assumption is certainly false: response rates depend on other factors , such as motivation and stimulus-specific properties ( e . g . , [15] ) . Nonetheless , this assumption enables the models considered below to make predictions about a wide range of experimental paradigms without getting bogged down in experimental minutiae . The same is true for many other computational models , and is helpful for making progress before more realistic theoretical assumptions can be refined . The Rescorla-Wagner model is the cornerstone of modern associative learning theory . While it has a number of crucial shortcomings [16] , the model stimulated decades of experimental research and served as the basis of more sophisticated models [17–19] . Learning is governed by the following equation: w n + 1 = w n + α x n δ n , ( 1 ) v n = w n ⊤ x n ( 2 ) where α ∈ [0 , 1] is a learning rate parameter ( also known as associability ) , δn = rn − vn is the prediction error , and vn is the reward expectation , which is taken to be monotonically related to the conditioned response . In the next section , I describe a probabilistic interpretation of this learning rule , which will play an important role in subsequent developments . I then discuss some empirical implications of the model . The probabilistic interpretation of the Rescorla-Wagner model given above shows that it is a maximum likelihood estimator of the weight vector . This estimator neglects the learner’s uncertainty by only representing the single most likely weight vector . Given that humans and other animals are able to report their uncertainty , and that these reports are often well-calibrated with veridical confidence ( i . e . , the probability of being correct; see [37] ) , it appears necessary to consider models that explicitly represent uncertainty . Moreover , such models are an important step towards understanding how the brain represents uncertainty [23 , 24] . Bayesian models of learning posit that the learner represents uncertainty in the form of a posterior distribution over hypotheses given data . In the case of associative learning , the posterior distribution is stipulated by Bayes’ rule as follows: p ( w n | x 1 : n ) ∝ p ( x 1 : n | w n ) p ( w n ) . ( 8 ) Under the LDS specified in Eqs 3–5 , the posterior is Gaussian with mean w ^ n and covariance matrix Σn , updated using the Kalman filter equations: w ^ n + 1 = w ^ n + k n δ n ( 9 ) Σ n + 1 = Σ n + τ 2 I - k n x n ⊤ ( Σ n + τ 2 I ) , ( 10 ) where w ^ 0 = 0 , Σ 0 = σ w 2 I , and kn is the Kalman gain: k n = ( Σ n + τ 2 I ) x n x n ⊤ ( Σ n + τ 2 I ) x n + σ r 2 . ( 11 ) Here the Kalman gain has replaced the learning rate α in the Rescorla-Wagner model . Importantly , the Kalman gain is stimulus-specific , dynamic and grows monotonically with the uncertainty encoded in the diagonals of the posterior covariance matrix Σn . This allows the Kalman filter model to explain some of the phenomena that are problematic for the Rescorla-Wagner model . Two factors govern the covariance matrix update . First , uncertainty grows over time due to the random diffusion of the weights ( Eq 4 ) ; this is expressed by the τ2I term in Eq 10 . The growth of uncertainty over time increases with the diffusion variance τ2 , leading to higher learning rates in more “volatile” environments . The relationship between volatility and learning rate follows intuitively from the fact that high volatility means that older information is less relevant and can therefore be forgotten [38 , 39] . The second factor governing the covariance matrix update is the reduction of uncertainty due to observation of data , as expressed by the term k n x n ⊤ ( Σ n + τ 2 I ) . Whenever a cue is observed , its variance on the diagonal of the covariance matrix is reduced , as are the covariances ( off-diagonals ) for any correlated cues . One implication of the Kalman filter is that repeated CS presentations will attenuate posterior uncertainty and therefore reduce the Kalman gain . As illustrated in Fig 2 , this reduction in gain produces latent inhibition , capturing the intuition that CS pre-exposure reduces “attention” ( associability or learning rate ) . The Kalman filter can also explain why interposing an interval between pre-exposure and conditioning attenuates latent inhibition [40]: The posterior variance grows over the interval ( due to random diffusion of the weights ) , increasing the Kalman gain . Thus , the Kalman filter can model some changes in learning that occur in the absence of prediction error , unlike the Rescorla-Wagner model . The Kalman filter can also account for the effects of various post-training manipulations , such as backward blocking [3 , 6] . During the compound training phase , the model learns that the cue weights must sum to 1 ( the reward value ) , and thus any weight configurations in which one weight is large necessitates that the other weight be small . Mathematically , this is encoded as negative covariance between the weights ( i . e . , the off-diagonals of Σn ) . As a consequence , learning that A predicts reward leads to a reduction in the associative strength for B . Beyond backward blocking , the Kalman filter can capture a wider range of recovery phenomena than has previously been simulated . Four examples are shown in Fig 3 ( see Methods for simulation details ) . As shown by Matzel and colleagues [34] , overshadowing ( AB→+ training leads to weaker responding to B compared to B→+ training ) can be counteracted by extinguishing one of the stimulus elements prior to test ( AB→+; A→- ) . Similarly , extinguishing the blocking stimulus in a forward blocking paradigm ( A→+; AB→+; A→-; B→ ? ) causes a recovery of responding to the blocked stimulus [35] , and extinguishing one of the stimulus A in an overexpectation paradigm ( A→+ / B→+; AB→+; A→-; B→ ? ) causes a recovery of responding to the other stimulus B [36] . Finally , extinguishing the excitatory stimulus A in a conditioned inhibition paradigm ( A→+ / AB→-; A→- ) reduces the negative associative strength of the inhibitory stimulus B [41] . All of these examples have a common structure shared with backward blocking , where compound training causes the acquisition of negative covariance between the stimulus elements . This negative covariance implies that post-training inflation or deflation of one stimulus will cause changes in beliefs about the other stimulus . Post-training recovery phenomena have inspired new theories that allow learning to occur for absent stimuli . For example , Van Hamme and Wasserman [18] developed an extension of the Rescorla-Wagner model in which the associative strengths for absent cues are modified just like present cues , but possibly with a smaller learning rate ( see also [19 , 42 , 43] ) . The Kalman filter provides a normative explanation of recovery phenomena , while retaining close similarities with classical theories like the Rescorla-Wagner model . The Kalman filter fixes some of the problems vexing the Rescorla-Wagner model , but a fundamental limitation remains: The Rescorla-Wagner model is a trial-level model , which means that it only makes predictions at the granularity of a trial , remaining blind to intra-trial structure such as stimulus duration and the inter-stimulus interval . While one can finesse this by treating each time-step in the model as a sub-division of a trial , such a solution is inadequate because it fails to capture the fact that conditioned responses are anticipatory of long-term future events . For example , interposing a delay between CS offset and US onset means that the CS never co-occurs with the US and hence should not produce any conditioning according to this particular real-time extension of the Rescorla-Wagner model ( contrary to the empirical data ) . It is possible to augment the Rescorla-Wagner model with a time-varying stimulus trace evoked by the CS , allowing the trace to enter into association with the US . This idea goes back to the work of Pavlov [26] and Hull [44] , who posited that the stimulus trace persists for several seconds following CS offset , decaying gradually over time . More complex stimulus traces have been explored by later researchers ( e . g . , [45 , 46] ) . While a persistent trace enables the model to capture aspects of intra-trial temporal structure , there is an additional problem: the association between the trace and the US can only be reinforced following US presentation , but contrary to this assumption it has been demonstrated empirically that an association can be reinforced without any pairing between the CS and US . As mentioned above , an example is second-order conditioning [26 , 30] , where A is paired with reward and subsequently B is paired with A , resulting in conditioned responding to B . An analogous phenomenon , known as conditioned reinforcement , has been studied in operant conditioning [47] . Somehow , a CS must be able to acquire the reinforcing properties of the US with which it has been paired . The TD model [9] offers a solution to both of these problems , grounded in a different rational analysis of associative learning . The underlying assumption of the TD model is that the associative learning system is designed to learn a prediction of long-term future reward , rather than immediate reward ( as was assumed in our rational analysis of the Rescorla-Wagner and Kalman filter models ) . Specifically , let us imagine an animal that traverses a “state space” defined by the configuration of stimuli , moving from xt at time t to xt+1 according to a transition distribution p ( xt+1∣xt ) . ( Note that we now index by t to emphasize that we are in “real time” ) . The value of state xt is defined as the expected discounted future return ( cumulative reward ) : V ( x t ) = E[ ∑ k = 0 ∞ γ k r t + k ] , ( 12 ) where γ ∈ [0 , 1] is a discount factor that controls how heavily the near future is weighted relative to the distant future . Applications of the TD model to associative learning assume that conditioned responding is monotonically related to the animal’s value estimate . This means that two stimuli might have the same expected reward , but responding will be higher to the stimulus that predicts greater cumulative reward in the future . The RL problem is to learn the value function . As is common in the RL literature [48 , 49] , I will assume that the value function can be approximated as a linear combination of stimuli: V ( x t ) = w t ⊤ x t . This reduces the RL problem to learning wt . This can be accomplished using an update very similar to that of the Rescorla-Wagner model [49]: w ^ t + 1 = w ^ t + α x t δ t , ( 13 ) where δt is now defined as the temporal difference prediction error: δ t = r t + γ w ^ t ⊤ x t + 1 - w ^ t ⊤ x t . ( 14 ) Except for the addition of the future reward expectation term γ w ^ t ⊤ x t + 1 , the TD prediction error is identical to the Rescorla-Wagner prediction error , and reduces to it when γ = 0 . In order to apply the TD model to associative learning tasks , it is necessary to specify a temporally extended stimulus representation . Sutton and Barto [9] adopted the complete serial compound ( CSC ) representation , which divides a stimulus into a sequence of non-overlapping bins . Thus , a stimulus lasting for two time steps would be represented by x1 = [1 , 0] and x2 = [0 , 1] . Although there are a number of problems with this representation [11 , 50–52] , I use it here for continuity with previous work . The TD model can account for a number of intra-trial phenomena , such as the effect of stimulus timing on acquisition and cue competition ( see [9 , 11] for extensive simulations ) . It also provides a natural explanation for second-order conditioning: despite the immediate reward term rt in Eq 14 being 0 for A→B trials , the future reward expectation term γ w ^ t ⊤ x t + 1 is positive ( due to the B→+ trials ) and hence the value of A is increased . In summary , the TD model has proven to be a successful real-time generalization of the Rescorla-Wagner model , and also has the advantage of being grounded in the normative theory of RL . However , it lacks the uncertainty-tracking mechanisms of the Kalman filter , which I argued are important for understanding CS pre-exposure and post-training recovery effects . I now turn to the problem of unifying the Kalman filter and TD models . Bayesian versions of TD learning have been developed in a number of different forms [13 , 53 , 54]; all of them have in common the idea that an agent tracks the entire distribution over discounted future returns , not just the mean . Of particular interest is Kalman TD , an elegant adaptation of the Kalman filtering machinery to TD learning developed by Geist and Pietquin [13] . Operationally , the only change from the Kalman filter model described above is to replace the stimulus features xn with their discounted time derivative , ht = γ xt + 1−xt . To see why this makes sense , note that the immediate reward can be expressed in terms of the difference between two values: r t= γ V ( x t + 1 ) - V ( x t ) = γ w t ⊤ x t + 1 - w t ⊤ x t =w t ⊤ ( γ x t + 1 - x t ) . ( 15 ) I have assumed here , as in the previous section , that values are linear in the stimulus features . As the derivation shows , this implies that rewards are linear in the discounted time derivative of the stimulus features . Under the assumption that the weights evolve over time as a Gaussian random walk and the rewards are corrupted by Gaussian noise , we can use the same LDS formulation described earlier , for which the Kalman filter implements Bayesian estimation . Kalman TD combines the strengths of Kalman filtering and TD learning: it is a real-time model that that represents a distribution over weights rather than a point estimate . These properties allow the model to capture both within-trial structure and retrospective revaluation . In the remainder of this section , I present several examples that illustrate the intersection of these phenomena , and compare the predictions of TD and Kalman TD ( since these examples involve within-trial structure , I do not consider the Kalman filter or Rescorla-Wagner ) . Denniston et al . [55] presented a series of experiments exploring recovery from overshadowing . In one experiment ( summarized in Fig 4A ) , the authors combined overshadowing and second-order conditioning to show that extinguishing an overshadowed stimulus allows its partner to better support second-order conditioning . Animals were divided into two groups , OV-A and OV-B . Both groups first learned to associate two light-tone compounds ( AX and BY ) with a US ( a footshock in this case ) . This compound training protocol was expected to result in overshadowing . One element of the compound was then extinguished ( A in group OV-A , B in group OV-B ) . Stimulus X was then used as a second-order reinforcer for conditioning of a novel stimulus , Z . Denniston et al . found that overshadowing reduced the ability of an overshadowed stimulus to support second-order conditioning , but this reduction could be attenuated if the overshadowing stimulus was extinguished . In particular , they found that responding at test to stimulus Z was greater in group OV-A than in group OV-B . Simulations show that KTD , but not TD , can capture this finding ( Fig 4B ) . While TD can capture second-order conditioning , it cannot explain why post-training extinction changes the value of an absent stimulus , because only the weights for presented stimuli are eligible for updating . The latter phenomenon is captured by the Kalman filter , which encodes the negative covariation between stimuli . As a consequence , the Kalman gain for stimulus X during Phase 2 ( despite X not appearing during this phase ) is negative , meaning that extinguishing A will cause inflation of X . By contrast , extinguishing B has no effect on the value of X , since B and X did not covary during Phase 1 . This is essentially the same logic that explains the post-training recovery phenomena described above , but applied to a second-order conditioning scenario outside the scope of the Kalman filter . One extensively studied aspect of second-order conditioning has been the effect of extinguishing the first-order stimulus on responding to the second-order stimulus . Rashotte and colleagues [56] reported a Pavlovian autoshaping experiment with pigeons in which extinction of the first-order stimulus reduces responding to the second-order stimulus . This finding has been replicated a number of times [57–59] , although notably it is not found in a number of other paradigms [30 , 60] , and a comprehensive explanation for this discrepancy is still lacking . Fig 5 shows that Kalman TD predicts sensitivity to first-order extinction , whereas TD predicts no sensitivity . The sensitivity of Kalman TD derives from the positive covariance between the first- and second-order stimuli , such that changes in the value of the first-order stimulus immediately affect the value of the second-order stimulus . I next turn to serial compound conditioning , which illustrates the within-trial behavior of Kalman TD . As summarized in Fig 6A , Gibbs et al . [61] studied the effects of extinguishing stimulus X following serial compound training ( Z→X→+ ) . They found that this extinction treatment reduced the conditioned response to Z ( see [15] for similar results ) . Kalman TD can account for this finding ( Fig 6B ) because the positive covariance between Z and X means that the value of Z is sensitive to post-training manipulations of X’s value ( Fig 6C ) . TD , which lacks a covariance-tracking mechanism , cannot account for this finding . In a second experiment ( Fig 7A ) , Gibbs et al . had the extinction phase occur prior to training , thereby making it a latent inhibition ( CS pre-exposure ) design . As with the extinction treatment , latent inhibition reduces responding to Z , a finding that can be accounted for by Kalman TD , but not TD ( Fig 7B ) . The Kalman TD account is essentially the same as the Kalman filter account of latent inhibition: Pre-exposure of X causes its posterior variance to decrease , which results in a concomitant reduction of the Kalman gain ( Fig 7C ) . A conceptually related design was studied by Shevill and Hall [62] . Instead of extinguishing the first-order stimulus X , they extinguished the second-order stimulus Z and examined the effect on responding to the first-order stimulus ( Fig 8A ) . This extinction procedure increased responding to the first-order stimulus relative to another first-order stimulus ( Y ) whose associated second-order stimulus had not been extinguished . This finding is predicted by Kalman TD , but not TD ( Fig 8B ) , because in a serial conditioning procedure the first-order stimulus overshadows the second-order stimulus , and extinguishing the first-order stimulus causes a recovery from overshadowing ( a reduced first-order value is evidence that the second-order stimulus was responsible for the outcome ) . Note that this explanation is essentially the same as the one provided by the Kalman filter for recovery from overshadowing with simultaneous compounds [34]; the key difference here is that in serial compounds the second-order stimulus tends to differentially overshadow the first-order stimulus [63] .
One of the important insights of the Pearce-Hall model [17] was that learning rate should increase with surprise—formalized as the absolute value of recent prediction errors . This model successfully predicts that inconsistently pairing a CS with an outcome enhances its learning rate in a subsequent training phase with a different outcome [68] . In the Kalman filter ( as well as in Kalman TD ) , changes in learning rate are driven solely by changes in the covariance matrix , which does not depend on outcomes . Thus , the model cannot explain any changes in learning rate that depend on prediction errors . One way to deal with this problem is to recognize that the animal may have uncertainty about the transition dynamics ( parameterized by τ ) , so that it learns simultaneously about the associative weights and τ . It is straightforward to show that the partial derivative of the log-likelihood with respect to τ monotonically increases with δ t 2 , which means that gradient ascent will increase τ when the squared prediction error is greater than 0 . This will give rise to qualitatively similar behavior to the Pearce-Hall model . Closely related Bayesian treatments have been recently explored , although not in the context of TD learning [38 , 39 , 69 , 70] . Another issue that arises in models of associative learning is the problem of feature ( or state space ) representation [71] . When we present an animal with a stimulus configuration , it is reasonable to expect that the animal applies some kind of processing to the stimulus representation . Some neural network models conceive this processing as the application of a non-linear transformation to the stimulus inputs , resulting in a hidden-layer representation that encodes configural features [64 , 72 , 73] . Other models derive stimulus representation from a clustering process that partitions stimulus inputs into a discrete set of states [7 , 71 , 74 , 75] . A related line of work has studied the representation of temporally extended stimuli; for example , a number of theories postulate a distributed representation of stimuli using basis functions with temporal receptive fields ( see [52] for a review ) . In general , any of these representations are compatible with Kalman TD as long as values are linear functions of the representation . While this may sound limiting , it is in fact extremely powerful , since any smooth function can be arbitrarily well approximated by a linear combination of suitably chosen basis functions [76] . The final issue I will mention here concerns instrumental learning: A complete theory of associative learning must account for associations between actions and outcomes . One influential framework for combining Pavlovian and instrumental learning processes is the actor-critic architecture [77] , according to which a Pavlovian “critic” learns state values , while an instrumental “actor” optimizes its policy using the critic’s prediction errors . Within this architecture , Kalman TD could function as a Bayesian critic . An interesting question that then arises is what role the critic’s uncertainty should play in guiding policy updating ( see [78] for one possibility ) . This paper makes several contributions . First , it provides a unifying review of several associative learning models , elucidating their connections and their grounding in normative computational principles . Second , it presents new simulations that highlight previously unappreciated aspects of these models . Third , it presents Kalman TD , a synthesis of these models . While this model has been described in other papers [13 , 14] , this is the first systematic application to associative learning . This paper demonstrates that several prominent themes in associative learning theory can be coherently unified . | How do we learn about associations between events ? The seminal Rescorla-Wagner model provided a simple yet powerful foundation for understanding associative learning . However , much subsequent research has uncovered fundamental limitations of the Rescorla-Wagner model . One response to these limitations has been to rethink associative learning from a normative statistical perspective: How would an ideal agent learn about associations ? First , an agent should track its uncertainty using Bayesian principles . Second , an agent should learn about long-term ( not just immediate ) reward , using reinforcement learning principles . This article brings together these principles into a single framework and shows how they synergistically account for a number of complex learning phenomena . | [
"Abstract",
"Introduction",
"Results",
"Discussion"
] | [] | 2015 | A Unifying Probabilistic View of Associative Learning |
The splice isoforms of vascular endothelial growth A ( VEGF ) each have different affinities for the extracellular matrix ( ECM ) and the coreceptor NRP1 , which leads to distinct vascular phenotypes in model systems expressing only a single VEGF isoform . ECM-immobilized VEGF can bind to and activate VEGF receptor 2 ( VEGFR2 ) directly , with a different pattern of site-specific phosphorylation than diffusible VEGF . To date , the way in which ECM binding alters the distribution of isoforms of VEGF and of the related placental growth factor ( PlGF ) in the body and resulting angiogenic signaling is not well-understood . Here , we extend our previous validated cell-level computational model of VEGFR2 ligation , intracellular trafficking , and site-specific phosphorylation , which captured differences in signaling by soluble and immobilized VEGF , to a multi-scale whole-body framework . This computational systems pharmacology model captures the ability of the ECM to regulate isoform-specific growth factor distribution distinctly for VEGF and PlGF , and to buffer free VEGF and PlGF levels in tissue . We show that binding of immobilized growth factor to VEGF receptors , both on endothelial cells and soluble VEGFR1 , is likely important to signaling in vivo . Additionally , our model predicts that VEGF isoform-specific properties lead to distinct profiles of VEGFR1 and VEGFR2 binding and VEGFR2 site-specific phosphorylation in vivo , mediated by Neuropilin-1 . These predicted signaling changes mirror those observed in murine systems expressing single VEGF isoforms . Simulations predict that , contrary to the ‘ligand-shifting hypothesis , ’ VEGF and PlGF do not compete for receptor binding at physiological concentrations , though PlGF is predicted to slightly increase VEGFR2 phosphorylation when over-expressed by 10-fold . These results are critical to design of appropriate therapeutic strategies to control VEGF availability and signaling in regenerative medicine applications .
The primary objectives of this study were: ( 1 ) to predict the distribution of VEGF and PlGF within the body , ( 2 ) to understand the effect of VEGF and PlGF on the balance of VEGFR1 and VEGFR2 ligation and VEGFR2 phosphorylation , ( 3 ) to quantify the effect of matrix-bound VEGF & PlGF binding to endothelial and soluble receptors on VEGFR signaling , and ( 4 ) to study the impact of changes in VEGF & PlGF isoform expression on absolute and relative VEGFR1 & VEGFR2 activation and site-specific phosphorylation of VEGFR2 , as a result of isoform-specific matrix- and NRP1-binding properties , all within the context of a healthy human body . The computational systems pharmacology model developed in this study is based on previously-developed computational models of VEGF distribution and receptor binding in vivo . These models have included VEGF165 , VEGF121 , VEGFR1 , VEGFR2 , soluble VEGFR1 ( sR1 ) , NRP1 , and sites in the interstitial matrix to which some growth factors and sR1 can bind [56–58] . The distribution of these proteins and their complexes has been examined in tissues of therapeutic interest ( healthy or PAD calf [58] , or tumor [59 , 60] ) , the blood , and non-diseased tissue ( main body mass ) [56 , 57] , in humans or mice [61 , 62] , incorporating transport between these compartments via vascular permeability and lymphatic drainage of tissues , and clearance of proteins from the plasma . By including multiple tissue compartments , we can compare quantities in a tissue of interest to those in the bulk of body tissue . In the present study , we greatly expand upon previous models to further capture the complexity of VEGF distribution and VEGF receptor activation in the body . For the first time , we include two isoforms of placental growth factor ( PlGF1 & PlGF2 ) , and the VEGF isoform VEGF189 . Additionally , we account for binding of matrix-immobilized ligands in the endothelial basement membrane ( EBM ) to cell-surface receptors ( VEGFR1 & VEGFR2 ) , binding of immobilized ligands throughout the interstitial space to soluble sR1 , and the ability of sR1 , when sequestered in the interstitial matrix , to bind some VEGF isoforms . To capture these effects , we simulate receptor trafficking and VEGFR2 tyrosine site-specific phosphorylation following ligand binding or unbinding explicitly , implementing the reactions in a previously-developed in vitro computational model that captures differences in VEGFR2 phosphorylation following stimulation with soluble or matrix-bound VEGF165 [20] . Finally , we leverage recent measurements to update endothelial cell surface receptor densities [63] .
To capture the pharmacokinetics of VEGF , PlGF , and sR1 distribution in the human body , we divide the body into three compartments: a healthy calf muscle ( gastrocnemius + soleus muscles ) , blood , and the main body mass ( the rest of the tissues ) , approximated with the properties of skeletal muscle ( Fig 1A ) . Transport between compartments occurs via bi-directional vascular permeability and lymphatic drainage of tissues ( into the blood ) , while growth factors and sR1 are cleared from the blood ( via the liver and kidneys ) , using rates previously determined ( S10 Table ) . Each tissue compartment includes physiological proportions of interstitial space , extracellular matrix ( ECM ) , endothelium , other parenchymal cells , and basement membranes for both the endothelium and parenchyma ( endothelial- EBM , and parenchymal- PBM ) . Within each tissue , we incorporate molecularly-detailed pharmacodynamics , including secretion into the interstitial space of VEGF and PlGF by parenchymal cells and sR1 by endothelial cells . In the interstitium , these diffusible proteins can then bind to heparan sulfate proteoglycans ( HSPGs ) in the ECM and basement membranes ( see Fig 1B , S11 Table ) , bind to receptors on endothelial cells ( ECs ) , or be removed from the compartment via physiological transport processes ( Fig 1A ) . VEGF and PlGF isoforms have different affinities for matrix sites and for the coreceptor NRP1 , which are included ( Table 1 ) , to account for isoform-specific ligand distribution and receptor activation . On the surface of and within endothelial cells , we simulate binding of sR1 to NRP1 , binding of PlGF to VEGFR1 and/or NRP1 , and binding of VEGF to VEGFR1 , VEGFR2 , and/or NRP1 , based on the binding properties of each protein ( summarized in Tables 1–3 and Fig 1E ) . Endothelial cell surface receptors are continually produced , internalized , recycled , and degraded , with trafficking rates that depend on ligation status and complex formation with NRP1 ( Fig 1C ) . We include detailed VEGFR2 trafficking based on a previous in vitro computational model ( S8 Table ) . Surface receptor production rates were tuned to match experimental measurements of cell surface receptor levels in human umbilical vein endothelial cells ( Table 4 ) . We also explicitly include phosphorylation and site-specific dephosphorylation of VEGFR2 ( Fig 1D ) , which is dependent on receptor trafficking , with higher net activation at Y1214 than Y1175 on the cell surface , and higher Y1175 phosphorylation in early ( Rab4/5 ) endosomes ( S9 Table ) , as a result of differential dephosphorylation of Y1175 and Y1214 on the cell surface and in early endosomes [20] . This allows us to study phosphorylation explicitly , instead of using receptor occupancy as a surrogate , and to look at relative activation of downstream signaling pathways leading to proliferation ( pY1175 via ERK1/2 ) and migration ( pY1214 via p38 ) . Due to the spatially-averaged nature of this model , gradients and heterogeneity in growth factor , soluble receptor , and cell surface receptor patterning are neglected . Instead , we examine the tissue-averaged behavior within the context of the human body . We neglect secretion of sR1 directly into the bloodstream , receptors present on the luminal side of ECs , and degradation of growth factors by proteases . All parameters are based on or fit to experimental data , either newly here or previously for other computational models . By building on previous modeling efforts , we have built more molecular detail into our models , while adding only a modest number of new parameters ( indicated in bold in Tables 1–4 and S7 Table ) . To simulate the time-course of each molecular species in each tissue and the blood , this model includes 635 nonlinear ordinary differential equations that are solved simultaneously . The model equations can be found in S1 Equations . The full set of differential equations was solved in Fortran using the Livermore Solver for Ordinary Differential Equations with Automatic method switching for stiff and nonstiff problems ( LSODA ) , on a laptop PC , with a relative error tolerance of 10−6 .
The ligand secretion and receptor production rates necessary to hit baseline ( healthy ) targets had to be fit simultaneously , due to the highly non-linear nature of the system . At our baseline steady-state , the VEGF production rate is 0 . 2830 molecules/myonuclear domain/s , the PlGF production rate is 0 . 0146 molecules/myonuclear domain/s , and the sR1 production rate is 0 . 0893 molecules/EC/s ( see Table 4 ) . The VEGF and sR1 production rates here are higher than previous estimates . This is unsurprising , given the changes in receptor levels , trafficking , and growth factor isoforms . Surprisingly , the PlGF production rate is lower than that for VEGF , despite a higher target plasma level ( see Flux Analysis section for the mechanism by which this occurs ) . To illustrate the nonlinearity of our model , we perturbed each ligand secretion and receptor production rate slightly ( 2% ) , and examined changes in plasma ligand and tissue receptor levels . As shown in Fig 2A , plasma VEGF and tissue VEGFR2 are highly sensitive to changes in either VEGF secretion or VEGFR2 production in the main body mass , with changes of 11–25% per percent change in input . As VEGF levels increase , more VEGFR2 becomes occupied , internalized , and degraded , reducing VEGFR2 levels and decreasing VEGF consumption ( Fig 2B and S1 Fig ) . Similarly , as VEGFR2 production increases , more VEGF is bound to VEGFR2 , internalized , and degraded , reducing VEGF levels and thus increasing EC surface VEGFR2 . This super-sensitivity was not present in previous models , where surface VEGFR2 levels were fixed ( see S1 Fig ) . This new , emergent result suggests that , lacking upregulation of VEGFR2 in response to VEGF , VEGFR2 levels would be highly sensitive to even small fluctuations in local VEGF concentration ( Fig 2 ) , highlighting the importance of dynamic adjustments to ligand and receptor expression in vivo . In the calf muscle , perturbing VEGFR2 production has a large impact on EC surface VEGFR2 , but little effect on plasma VEGF , due to the smaller size of the compartment . Changes in receptor production in one tissue compartment have little effect on receptor levels in the other tissue compartment . In this model , we assume the same rates for ligand production in both the healthy calf muscle and the main body mass . As such , perturbing the VEGF secretion rate ( in both compartments ) alters the receptor levels in both tissues ( Fig 2 ) . Due to differences in the geometric parameterizations of the calf and other tissues ( S7 Table ) , using the same ligand secretion rates results in different interstitial VEGF , sR1 , and PlGF levels ( Fig 3D ) . We focus primarily on quantities measured in the “Main Body Mass” compartment , which , due to its larger size , represents the primary determinant of plasma VEGF , sR1 , and PlGF levels . After establishing the secretion and production rates required to achieve basal targets , we next examined the steady-state distribution of VEGF , PlGF , and sR1 . Having examined the distribution of VEGF , PlGF , and sR1 , we next zoomed in to examine the effect of these proteins and their distributions on the binding and activation of endothelial VEGFR1 and VEGFR2 within healthy tissue . It is clear that the different proteins—ligands , soluble receptors , and co-receptors—regulating VEGFR1 and VEGFR2 activation do not act in isolation . Changes to any single feature affect the total multi-factor system in a way that is difficult to predict without the use of a computational model . Here , we perturb several interactions that are of interest therapeutically , and/or are included in this model for the first time . The most convincing evidence to date of differential signaling by VEGF isoforms is the distinct vascular phenotypes of mice or human tumors ( implanted in mice ) expressing only single isoforms of VEGF , with VEGF121-only tissues producing high diameter , sparsely branched networks , VEGF165-only tissue a relatively normal phenotype , and VEGF189-only tissues networks of thin , highly branched vessels . Endothelial cells isolated from these single isoform-expressing mice also display distinct signaling and behavior in cell culture [81] . It is assumed that similar regulation occurs in humans . To better understand VEGF isoform-specific signaling in the context of the human , as well as to qualitatively validate our model , we simulated expression of a single VEGF isoform in the human body . While no significant changes in VEGFR1 or VEGFR2 mRNA were observed in the muscle of mice expressing only VEGF120 [82] ( equivalent to human VEGF121 ) , we re-fit our model for each case , in order to maintain target ligand and receptor levels ( S13 Table ) . The need for these changes in receptor production and ligand secretion rates may be a result of differences between humans and mice , or underlying compensation mechanisms and physiological changes in the engineered mice [82] not included in this model . Consistent with observations in mice , ligand distribution and VEGFR2 activation are more similar to wild type ( baseline ) in the VEGF165-only than the VEGF121-only or VEGF189-only cases ( Fig 8A and 8B ) . Similar to the baseline case ( Fig 5 ) , where all three isoforms are expressed , with single VEGF isoform expression the ratio of migratory to proliferative signaling downstream of VEGFR2 ( pY1214/pY1175 ) is predicted to increase with isoform length , paralleling the observed phenotypes ( Fig 8C ) . The model’s ability to capture this trend provides qualitative validation of our isoform-specific signaling predictions in vivo . Interestingly , the model also predicts other changes , in free VEGF levels in tissue interstitium ( Fig 8A ) and in relative activation of VEGFR1 and VEGFR2 ( Fig 8B and 8D , S1 File ) .
Our model predicts that , based on their binding properties and in vivo concentrations , PlGF and VEGF have distinct distributions within the body . PlGF2 , binding to the ECM more strongly than VEGF , is bound to interstitial matrix sites at very high levels ( ~1 nM in tissue: soluble + ECM-bound + EC-bound predicted , Fig 3C ) , forming a large reservoir available for proteolytic release . Despite high tissue PlGF levels , our simulations predict that only about 30% of ligated EC surface VEGFR1 is bound to PlGF . As a result , while most VEGF removal from tissue is predicted to occur via binding to endothelial receptors , only 25% of PlGF was predicted to bind to and be subsequently degraded by endothelial cells . PlGF also binds VEGFR1 on other cells , e . g . monocytes and macrophages , that are implicated in arteriogenesis [26 , 83] . We found that removing PlGF or increasing PlGF secretion has only a modest effect on predicted VEGFR2 phosphorylation , while substantially altering VEGFR1 activation ( Fig 6A ) . This result suggests that observed physiological PlGF-dependent pro-angiogenic effects are likely mediated directly by VEGFR1 , either on ECs or other cells , and not via changes in VEGFR2 signaling , contrary to the ‘ligand-shifting hypothesis’ . This result implicates VEGFR1 in the impaired angiogenic responses to ischemia , wound healing , and cancer [21] observed in mice lacking PlGF . It also implicates VEGFR1 in diseases where PlGF levels are known to change or to be predictive of prognosis , e . g . pre-eclampsia [42] and breast cancer [84] . The pro-angiogenic effects of PlGF likely also rely on its ability to up-regulate other growth factors , including VEGF , FGF2 , and PDGF [85 , 86] . This result is not inconsistent with recent work by the Alitalo group showing that therapeutic over-expression of VEGFB ( which like PlGF binds only VEGFR1 ) in mice improves metabolic health even following endothelial Flt1 gene deletion , and inhibits doxorubicin-induced cardiotoxicity [54 , 87] . Competition between ligands is concentration-dependent , and in these studies , VEGFB protein levels were elevated 20-fold or more in serum , heart , liver , and white adipose tissue . Our model predicts that competition is not a driver of PlGF signaling in physiological conditions , but does not preclude the existence of competition following supraphysiologic therapy . Indeed , at 10-fold PlGF over-expression , outside of the concentration range likely to be observed in untreated healthy or diseased tissue [42] , the model does begin to predict an effect on VEGFR2 signaling . Both the ECM and sR1 regulate tissue levels of free interstitial VEGF and PlGF , the amount of growth factor available to bind ECs , and the steady-state distribution of ligand throughout the body ( Fig 3 ) . The model predicts that sR1 modulates the magnitude of EC receptor ligation , potentially also altering the balance of signaling via VEGFR1 vs . VEGFR2 ( Fig 6G ) . This is of therapeutic interest because ratios of VEGF or PlGF to sR1 levels in plasma are increasingly of interest as a biomarker ( e . g . in pre-eclampsia ) [70] , and sR1 levels increase in diabetic mice following hindlimb ischemia [88] . Including binding of immobilized ligands to sR1 , and binding of immobilized sR1 to VEGF121 and PlGF1 , increases total extracellular VEGF and PlGF stored in tissue ( Fig 7 ) . While there is not yet evidence to prove the existence of such complexes , the heparin- and ligand-binding sites on sR1 are distinct , as are the heparin- and receptor-binding domains on VEGF and PlGF , and therefore these complexes are likely . Unlike matrix-ligand-sR1 complexes , VEGF immobilized to both surfaces and ECM proteins has been shown to bind and activate VEGFR2 in vitro , preferentially increasing VEGFR2 activation of tyrosine Y1214 , upstream of p38 phosphorylation and migratory cell behavior , demonstrating an important role for physical immobilization of VEGF in signal regulation in vitro [18 , 19 , 89] . However , whether VEGFR2 ligation by immobilized VEGF would occur to any notable extent in vivo , and what the physiological impact on EC receptor signaling would be , have been unknown . Here , we saw that including these reactions increased EC receptor ligation and altered VEGFR2 signaling ( Fig 7 ) . While the number of available sites in the EBM is not well-established , our model suggests that these M-L-R complexes may make up a small but significant portion of ligated EC receptors ( Fig 4D ) . To improve our estimates of the extent of EC receptor ligand by EBM-bound growth factor , it is necessary to obtain better estimates of heparin-binding sites in basement membranes . Interestingly , the fraction of ligated VEGFR1 bound to immobilized ligand was predicted to be higher than that for VEGFR2 , owing largely to the strong M-PlGF2 affinity ( Fig 6F ) . To date , the impact of VEGFR1 ligation by immobilized ligand has not been studied . However , as these are largely PlGF2-VEGFR1 complexes ( Fig 6F ) , EBM binding site density may shift relative ligation of VEGFR1 by VEGF versus PlGF , which is known to alter VEGFR1-mediated signaling [44] . Spatial patterning of receptor ligation by soluble and immobilized ligand is also likely to be important , but cannot be examined with this model . Additionally , the potential roles for HSPGs and NRP1 expressed on other cells engaging with VEGFR2 in trans [90 , 91] are of interest for future study . We were interested in differences in signaling between VEGF isoforms upon binding to VEGFR1 and VEGFR2 . Explicitly simulating VEGFR2 trafficking and site-specific phosphorylation , placed in the context of physiological geometry and transport processes , allowed us to predict isoform-specific VEGFR2 signaling in vivo ( Fig 5 ) . Immobilization in the matrix alters VEGF distribution and the resulting signaling , while NRP1 alters VEGF-receptor binding and trafficking . By including these isoform-specific properties , the model predicts that VEGF121 induces a shift in VEGFR2 distribution towards early signaling endosomes , decreasing the signaling ratio pY1214/pY1175 , and shifting the net cellular signaling towards proliferation . Conversely , a larger portion of VEGFR2 bound to VEGF189 was localized on the EC surface at steady-state , increasing pY1214/pY1175 , and shifting the balance towards pro-migratory signaling ( Fig 5C ) . This isoform-specific patterning in VEGFR2 signaling was seen in both the baseline case ( Fig 5C ) , with all three VEGF isoforms present , and in cases where only single isoforms of VEGF were expressed ( Fig 8C ) . This is key validation , as our simulated signaling predictions in humans match the observed vascular phenotypes in mice or tumors expressing single VEGF isoforms ( Fig 9B ) . Interestingly , in the single isoform cases , change in relative activation of VEGFR1 and VEGFR2 were also predicted ( Fig 8B ) , which may contribute to these phenotypes [92 , 93] . This is in line with another interesting model prediction; while all VEGF isoforms can bind to both VEGFR1 and VEGFR2 , physiologically it appears that VEGF165 and VEGF189 bind almost exclusively to VEGFR2 , while VEGF121 comprises a large portion of the ligand on VEGFR1 , and also binds VEGFR2 to an extent ( Fig 4D ) . This segregation of ligands suggests that , while ligand levels are limiting for receptor binding , VEGFR1 and VEGFR2 don’t directly compete for VEGF in vivo , instead binding to largely distinct subsets of ligands dictated primarily by isoform-specific NRP1-binding properties ( Fig 9C ) . The relative levels of VEGF isoforms are not yet extensively-characterized , but they are known to vary by tissue and to change in disease [69 , 82 , 94 , 95] . As such , this model can be used to understand splicing-induced tissue- and disease-specific changes in VEGF receptor signaling . Our model is built upon experimental data and a validated model of VEGFR2 signaling in vitro , and provides new insight into distribution of and signaling by VEGF and PlGF isoforms in vivo . However , when interpreting the results , it is important to acknowledge mismatch between model predictions and experimental measurements , which may result from limitations of our modeling approach , uncertainly in interpretation of experimental measures , and/or missing understanding of underlying biological mechanism . Similar to previous models , our predicted interstitial VEGF concentrations when fitting the model to measured plasma VEGF levels are higher than those measured in tissues using microdialysis . This discrepancy could be due to: difficulty in obtaining accurate measurements for high molecular weight proteins using microdialysis; production of VEGF by blood sources ( e . g . PBMCs , platelets ) or specific organs ( e . g . highly fenestrated tissue ) , reducing the requisite VEGF production by skeletal muscle; or degradation of VEGF by tissue-resident proteases and/or other cell types expressing VEGF receptors ( modeled in [96 , 97] ) . Inclusion of proteases in the model would reduce immobilized growth factor stores at steady state . Additionally , as in previous models , the predicted fraction of plasma sR1 bound to ligand was higher than the experimentally-measured fraction . There are other soluble receptors that may be important to consider and are not included here . There may also be limitations with the experimental method that make these in vivo measurements inaccurate . To quantify the importance of some difficult-to-measure parameters , as well as reactions included in this model for the first time ( some of which have not been explicitly demonstrated experimentally ) , we analyzed the sensitivity of many new or poorly characterized parameters ( see S3 Fig and S1 File ) . In order to achieve simulation at the whole body scale , compartment models neglect spatial effects , instead predicting only average values for tissue . The interstitial space of the tissue , the cell surface of endothelial cells and the cell surface of myocytes are still independent entities in this case and each is treated as well-mixed . Detailed study of gradients in interstitial space and along cell surfaces , which are difficult to measure in vivo but are likely key to angiogenic signaling , requires development of detailed 2- and 3-dimensional models of tissue and experimental set-ups , calibrated to match predicted average concentrations from compartment models [98–101] such as the one presented here . Much work remains to fully understand the role of spatial gradients of VEGF distribution and receptor activation in health , disease , and response to therapy . This model integrates detailed regulation of VEGF and PlGF distribution and binding to EC VEGFR1 and VEGFR2 by sR1 , the ECM , and NRP1 into a multi-scale pharmacokinetic/pharmacodynamic ( PK/PD ) framework . The resulting model predicts that all of these features interact , and contribute to regulation of tissue-level VEGF family signaling . While many model predictions are difficult to validate in vivo , the mechanisms included were first modeled using detailed in vitro measurements , and validated in many cases on the cellular level , before being put in a physiological context using an existing PK/PD framework . By progressively adding complexity , we can study the impact of each contribution , and compare simulation results to quantities that are measurable and to observable phenotypes , such as the vascular morphologies in mice expressing single isoforms of VEGF . By the same turn , this model provides a window into details of growth factor distribution and signaling that are essentially impossible to measure ( especially on the protein level ) , though in many cases implicated in disease-related impairment in angiogenic response , or targeted by potential therapies . The lack of approved pro-angiogenic therapies to date makes it clear that a better understanding of the molecular mechanisms driving disease is critical to identify more effective drug targets , optimize drug properties ( e . g . affinity ) , and avoid off-target effects leading to toxicity and drug failure [55] . This work can be extended to disease applications with changes in VEGF splicing , and to compare results in humans versus mice , to aid in translation of therapeutics targeting the VEGF system and to further validate the model against data obtained in mice . | Angiogenesis , the growth of new blood vessels from the existing vasculature , is critical for maintenance of health and response to injury . In ischemic disease , this process is impaired , but therapies targeting a key family of proteins , the vascular endothelial growth factors ( VEGF ) , have failed to translate clinically . This suggests a need for deeper understanding of the complex regulation underlying angiogenic signaling . Here , we translate a previously developed and validated model of VEGF family signaling into a human , whole-body framework . The different splice isoforms of VEGF and the related PlGF proteins have different affinities for the extracellular matrix ( ECM ) and the co-receptor Neuropilin-1 . Using our model , we examine the effect of these different binding properties on the distribution of each isoform in tissue , and subsequent receptor signaling . The model predicts isoform-specific receptor activation that is consistent with observed vascular phenotypes in mice expressing a single VEGF isoform; non-ECM-binding isoforms lead to signaling that promotes cell proliferation , while strong ECM-binding promotes migratory signaling and increased vessel branching . This understanding is critical for design of biomaterials that manipulate VEGF-ECM binding to control growth factor delivery , and for understanding of splicing-induced changes in VEGF family signaling in different tissues and in disease . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"cell",
"binding",
"phosphorylation",
"cell",
"physiology",
"medicine",
"and",
"health",
"sciences",
"placental",
"growth",
"factor",
"body",
"fluids",
"vegf",
"signaling",
"endothelial",
"cells",
"epithelial",
"cells",
"endocrine",
"physiology",
"growth",
"factors",
"cellular",
"structures",
"and",
"organelles",
"animal",
"cells",
"proteins",
"extracellular",
"matrix",
"endocrinology",
"biological",
"tissue",
"blood",
"plasma",
"biochemistry",
"signal",
"transduction",
"blood",
"cell",
"biology",
"post-translational",
"modification",
"anatomy",
"physiology",
"epithelium",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"cell",
"signaling"
] | 2017 | A computational analysis of in vivo VEGFR activation by multiple co-expressed ligands |
Although Plasmodium vivax is responsible for the majority of malaria infections outside Africa , little is known about its evolution and pathway to humans . Its closest genetic relative , P . vivax-like , was discovered in African great apes and is hypothesized to have given rise to P . vivax in humans . To unravel the evolutionary history and adaptation of P . vivax to different host environments , we generated using long- and short-read sequence technologies 2 new P . vivax-like reference genomes and 9 additional P . vivax-like genotypes . Analyses show that the genomes of P . vivax and P . vivax-like are highly similar and colinear within the core regions . Phylogenetic analyses clearly show that P . vivax-like parasites form a genetically distinct clade from P . vivax . Concerning the relative divergence dating , we show that the evolution of P . vivax in humans did not occur at the same time as the other agents of human malaria , thus suggesting that the transfer of Plasmodium parasites to humans happened several times independently over the history of the Homo genus . We further identify several key genes that exhibit signatures of positive selection exclusively in the human P . vivax parasites . Two of these genes have been identified to also be under positive selection in the other main human malaria agent , P . falciparum , thus suggesting their key role in the evolution of the ability of these parasites to infect humans or their anthropophilic vectors . Finally , we demonstrate that some gene families important for red blood cell ( RBC ) invasion ( a key step of the life cycle of these parasites ) have undergone lineage-specific evolution in the human parasite ( e . g . , reticulocyte-binding proteins [RBPs] ) .
Eleven P . vivax-like genotypes were obtained from 2 different kinds of samples: 10 infected chimpanzee blood samples collected during successive routine sanitary controls of chimpanzees living in the Park of La Lékédi ( a sanctuary in Gabon ) and 1 infected Anopheles mosquito ( An . moucheti ) collected during an entomological survey carried out in the same park ( S1 Table ) [11] . For blood samples , white blood cells were depleted using the CF11 method [12] to reduce the amount of host DNA . After DNA extraction , samples were subjected to whole-genome amplification ( WGA ) in order to obtain sufficient parasite DNA for library preparation . Sequencing was then performed using short-read Illumina technology . For one sample ( Pvl06 ) , long-read sequencing ( Pacific Biosciences [PacBio] technology ) was performed in order to get a better coverage of regions containing subtelomeric gene families . Among the 11 samples , 10 presented mixed infections with other Plasmodium species ( S1 and S2 Tables ) . Four samples containing P . gaboni or P . malariae-like co-infections were used in other studies ( see S1 Table ) [13 , 14] . In order to obtain the P . vivax-like genotypes and to preclude errors due to co-infections with other Plasmodium species , sequencing reads were extracted based on their similarity to the reference genome sequence of P . vivax , PvP01 [15] , and any reads mapping against a reference genome of a Laverania chimpanzee-infecting species ( e . g . , PrG01 , PbilcG01 , and PGAB01; see Otto and colleagues [13] ) were removed ( S2 Table ) . Concerning multiple infections with several P . vivax-like strains , only 2 ( Pvl09 and Pvl10 ) seemed to be multiply infected as suggested by the reference allele frequency ( RAF ) distributions ( see Materials and methods section and S1 Fig ) . In order to avoid any bias in the genomic analysis due to multiple infections with several strains of P . vivax-like , only 1 variant was extracted per sample . This was done by considering the allele with the highest frequency in the calling and filtering analysis of Single Nucleotide Variants ( SNVs ) ( see Materials and methods section ) . Sequencing reads from 2 samples , one obtained using Illumina sequencing , Pvl01 , and another using PacBio technology , Pvl06 , were used to perform de novo genome assemblies and were annotated to produce reference genomes for P . vivax-like ( S1 Table ) . Of the 2 assemblies , Pvl01 is of considerably higher quality ( 4 , 570 one-to-one orthologues to the PvP01 reference genome compared with 2 , 925 for Pvl06 [Table 1] ) . Both assemblies consist of 14 supercontigs ( corresponding to the 14 P . vivax chromosomes ) —and 1 , 176 and 351 unassigned contigs—comprising a total of 27 . 5 Mb and 18 . 8 Mb in size , respectively , for Pvl01 and Pvl06 , respectively . After annotation with Companion [16] , these 2 genomes contained 5 , 532 and 4 , 953 annotated genes ( Table 1 ) . We obtained for the 9 remaining samples between 2 . 9 and 86 million of reads that paired with the P . vivax PvP01 reference genome ( mean ± SD = 2 . 17 ± 25 . 71 ) , with a mean depth per high-quality position ranging from 13 . 98 to 335 . 1 ( mean ± SD = 93 . 43 ± 99 . 51; see S3 Table ) . These genome sequences were used for SNV calling for population genetic and phylogenetic analyses .
Comparing the P . vivax-like reference genomes to those of P . vivax ( PvP01 and SalI ) [3 , 15] , P . cynomolgi ( B and M strains ) [8 , 17] , and P . knowlesi ( H strain ) [18] reveals several similarities , including a similar guanine–cytosine ( GC ) content and extensive collinearity and conservation of gene content and organization ( Table 1 ) . The P . vivax-like core genome sequences are completely syntenic to the P . vivax PvP01 reference genome sequence ( S2 and S3 Figs ) . Because multigene families are known to evolve extremely rapidly in their genome structure , obtaining the full genomes of species closer to human P . vivax is fundamental for a better understanding of its evolution , adaptation , and emergence in different host species . For Plasmodium parasites , most species-specific genes are part of large gene families , such as var genes in P . falciparum or pir genes that are present in all Plasmodium genomes studied [19 , 20] . Table 2 provides a summarized view of gene content and copy number of the main multigene families in P . vivax-like in comparison with P . vivax , P . knowlesi , and P . cynomolgi . Even if certain subtelomeric regions of our reference genomes ( S2 and S3 Figs ) are not complete , at least one copy of each major gene family was detected ( Table 2 ) . In comparison with P . vivax ( and expected because of the partial subtelomeric sequencing coverage ) , the number of copies in each family was generally lower or equal in P . vivax-like . For these families , all genes were functional except for the Cytoadherence-linked asexual gene ( clag ) family . For the clag family , all genes are functional except the one situated on chromosome 8 for P . vivax-like ( confirmed for both Pvl01 and Pvl06 ) ( S4 Fig ) . The clag family , strictly conserved in malaria parasites , is an essential gene family in host–parasite interactions , playing a role in merozoite invasion , parasitophorous vacuole formation , and in the uptake of ions and nutrients from the host plasma [21 , 22] . The pseudogenization of the clag gene on chromosome 8 for P . vivax-like suggests that this species lost the clag gene during its adaptation to the ape host . During the life cycle of Plasmodium parasites , host RBC invasion is mediated by specific interactions between parasite ligands and host erythrocyte receptors . Two major multigene families are involved in RBC invasion: the Duffy-binding proteins ( dbp ) and the reticulocyte-binding proteins ( rbp ) multigene families [23] . DBP is a protein secreted by the micronemes of the merozoite stage that binds to the Duffy Antigen Receptor for Chemokine ( DARC ) to invade RBCs . P . vivax is characterized in its genome by 2 dbp genes ( dbp1 on chromosome 6 and dbp2 on chromosome 1 ) that seem to be essential for RBC invasion , as demonstrated by their inability to infect individuals not expressing the Duffy receptor on the surface of their RBCs ( i . e . , Duffy-negative individuals ) [24–26] . In the reference genomes of P . vivax-like ( Pvl01 and Pvl06 ) , we observe the dbp1 gene as in the P . vivax genome PvP01 ( Table 2 ) and also in the other Plasmodium species ( P . knowlesi and P . cynomolgi ) ; however , we did not observe the dbp2 gene ( no read obtained mapping to this region ) . This observation was confirmed in the other genotypes sequenced in this study . Knowing that gorillas and chimpanzees are all described as Duffy positive [10] , we propose that P . vivax-like parasites infect only Duffy-positive hosts , which could be associated with the absence of the dbp2 gene . This would be in accordance with the fact that the only described transfer of P . vivax-like to humans was in a Caucasian Duffy-positive individual [9] and that no transfers of P . vivax-like have been recorded in Central African Duffy-negative populations despite the fact that they live in close proximity to infected ape populations [27] . rbp genes encode a merozoite surface protein family present across all Plasmodium species and known to be involved in RBC invasion and host specificity [23] . Among rbp family genes , 3 gene classes ( rbp1 , rbp2 , and rbp3 ) exist and are associated with the ability of Plasmodium parasites to invade different maturation stages of RBCs . In this study , comparison of the organization and characteristics of the rbp gene family between P . vivax , P . vivax-like , P . knowlesi , and P . cynomolgi ( Fig 1 and Table 2 ) first reveals that gene classes RBP2 and RBP3 are ancestral to the divergence of all these species except P . knowlesi . Second , an expansion of the rbp2 class is observed in the P . vivax/P . vivax-like/P . cynomolgi lineage ( Fig 1A ) , suggesting that , in this lineage , specific expansion likely occurred during the evolution of these species . Finally , rbp3 genes , which are supposed to confer the ability to infect normocytes , are functional in all species except in P . vivax ( for which the gene is pseudogenized in both SalI and PvP01 strains ) , suggesting that P . vivax lost the ability to infect normocytes or has developed an ability to infect specifically only reticulocytes during its adaptation to human RBCs ( Fig 1A and 1B ) .
Conservation of the gene content between P . vivax-like with the other primate-infective Plasmodium species has enabled us to reconstruct with confidence the relationships between the different species and to estimate the relative age of the different speciation events . This analysis confirmed the position of P . vivax-like as the closest sister lineage of P . vivax ( Fig 2 ) . Regarding the estimation of divergence times using genomic information , different methods were recently used for Plasmodium , such as the one implemented in Generalized Phylogenetic Coalescent Sampler ( G-PhoCS ) [28] or the one developed by Silva and colleagues [29] . G-PhoCS uses a Bayesian Markov Chain Monte Carlo ( MCMC ) approach to infer , based on the information provided by multiple loci , the divergence time between species . This method has been applied in 2 recent studies for Plasmodium parasites—one aiming at estimating the relative split times between the 2 P . ovale subspecies and between P . malariae and P . malariae-like [14] , the other to estimate the divergence time within the Laverania subgenus , a subgenus including P . falciparum and all its closest ape relatives [13] . The Silva method is based on the estimate of the sequence divergence in different proteins and comparison of this divergence measured between different lineages [29] . In this method , the regression slope of the divergence between the proteins in 2 lineages reflects their relative age . The advantage is that it does not rely on an estimate of mutation rate . Finally , it has already been used in a recent study estimating avian and primate Plasmodium species divergence times [30] . Here , without calibration points and a good estimation of the P . vivax and other Plasmodium species substitution rates , our aim was to evaluate the divergence time between P . vivax and P . vivax-like relative to the divergence time between the other primate–nonprimate Plasmodium species pairs . To evaluate the influence of the approach on the estimations , we used 2 strategies to estimate divergence time between P . vivax and P . vivax-like relative to the other divergence events within the tree: the Silva method [29] and the RelTime method ( see Materials and methods section for a description of the two methods ) [31 , 32] . We first used the Silva method [29] , checking for the influence of the model of evolution and the reference species pair on the analysis ( see Materials and methods section ) . Because neither influenced the results of the relative ages of species pairs ( see Fig 2 , S5 Fig and S4 Table ) , we only report the results for the Silva method using the JTT ( Jones , Taylor , and Thornton ) model of evolution and the P . ovale curtisi–P . ovale wallikeri pair as the species pair of reference as in Böhme and colleagues in Fig 2 [30] . While the 2 methods are mostly in agreement ( with RelTime showing larger confidence intervals [CIs] ) , they show a discrepancy concerning the relative divergence time of the P . malariae and P . malariae-like parasites ( see Fig 2 and S4 Table ) . However , both methods suggest that the evolution of P . vivax in humans did not occur at the same time as the other human malaria agents of the Plasmodium genus ( i . e . , P . malariae and P . falciparum ) . The analyses show that the time of the split between P . vivax and P . vivax-like happened before the divergence between P . falciparum and P . praefalciparum ( about 2 . 3 times earlier ) . It is unclear from our analyses whether the divergence between the 2 P . malariae species occurred before or after the divergence of the other 2 human–nonhuman parasite pairs . Unlike us , a previous study estimated , using the G-PhoCS method , the relative dating of the split between P . reichenowi ( a chimpanzee parasite ) and P . falciparum ( P . praefalciparum was not available at that time ) and that of the split between P . malariae and P . malariae-like ( note that similar estimates were recently obtained using another set of data using the Silva method ) [14] to have occurred at the same time . This discrepancy between methods suggests that the same strict molecular clock ( which is a hypothesis of the Silva method ) may not apply over the entire tree ( especially for the Laverania subgenus because of their extremely low GC content in comparison with other Plasmodium species ) . Whatever the method used , all these estimates nevertheless suggest that the evolution of P . vivax in humans did not occur at the same time as the other human malaria species and that the transfer of Plasmodium parasites to humans may have happened several times independently over the history of the Homo genus [33–37] .
To analyze the relationship between our 11 P . vivax-like isolates and human P . vivax , we completed our dataset with 19 published human P . vivax genomes ( S1 Table ) [38] . All sequencing reads were aligned against the PvP01 reference genome [15] , and SNVs were called and filtered as described in the Materials and methods section . Maximum likelihood phylogenetic trees were then produced based on 100 , 616 SNVs . Our results clearly demonstrate the presence of 2 significantly distinct genetic clades ( with a bootstrap value of 100 ) composed of P . vivax-like strains on one side and human P . vivax isolates on the other side ( Fig 3 ) . This result differs from previous results suggesting that human strains formed a monophyletic clade within the radiation of ape P . vivax-like parasites [10] . One explanation for this difference with previous published results could be that it is due to a phenomenon called Incomplete Lineage Sorting ( ILS ) or to a lack of phylogenetic signal for phylogenies performed on a single or few genes . ILS is the discordance observed between some gene trees and the species or population tree due to the coalescence of gene copies in an ancestral species or population [39] . Such a phenomenon is often observed when species or population divergence is recent , which is the case for P . vivax/P . vivax-like [40 , 41] . ILS may thus result in the wrong conclusion of P . vivax and P . vivax-like populations being intermixed and P . vivax diversity being included in the diversity of P . vivax-like . A lack of phylogenetic signal , which occurs frequently when species diverged recently , would have similar consequences . In order to test the implication of ILS or lack of phylogenetic signal , we generated a phylogenetic tree and a reticulated network on partial mitochondrial genomes of P . vivax and P . vivax-like obtained in Liu and colleagues [10] , Prugnolle and colleagues [9] , and in the current study . These analyses show that partial mitochondrial genetic information is not enough to make a final conclusion on the origin of P . vivax parasites ( S6 Fig ) . When considering the polymorphism level of this portion of mitochondrial genomes , over the 2 , 483 bp , only 127 positions showed variability , 20 of them being a position specific to the outgroup P . cynomolgi . Among the 107 variable positions found in the P . vivax and P . vivax-like samples , 87 were singletons , meaning that only 20 SNVs were shared by more than 2 individuals . This suggests that the discrepancy between our phylogeny and the one of Liu and colleagues [10] is probably because of a lack of phylogenetic signal . Indeed , in our study , the use of significantly more genetic information from throughout the genome , both in genic and intergenic regions , provides a more accurate picture of the genetic relationships between the different parasite species . Reducing our genetic data to single genes ( as performed in previous studies ) or a limited number of SNVs also generates phylogenies in which P . vivax is included within the diversity of P . vivax-like ( see S7 Fig ) . Another explanation for the discrepancy between our results and the results from Liu and colleagues [10] could be that we are missing part of the diversity of P . vivax-like given that we obtained the genomes from a limited number of parasites isolated from a small population of apes in Gabon . Nevertheless , this hypothesis does not seem to hold in light of the tree and network produced using all the available mitochondrial sequences of P . vivax-like and our data , as isolates are distributed all over the currently known genetic diversity of P . vivax-like ( S6 Fig ) . Our results show that P . vivax-like is composed of 2 distinct lineages: one including the 2 reference genomes ( Pvl01 and Pvl06 ) and 7 other isolates that will hereafter be referred to as P . vivax-like 1 , and another one including 2 isolates ( Pvl09 and Pvl10 ) that will hereafter be referred to as P . vivax-like 2 ( Fig 3 ) . This sub-structuration of P . vivax-like is confirmed in the mitochondrial-reticulated network obtained using a larger number of isolates ( see S6 Fig ) . These 2 lineages may thus reflect an ancient split within P . vivax-like or be the consequence of a recent introgression or hybridization event between P . vivax-like and P . vivax in Africa . A search of recent recombination events between lineages using SplitsTree ( http://www . splitstree . org/ ) [77] does not support this latter hypothesis ( S6 Fig ) . Previous studies highlighted the high genetic diversity of P . vivax-like populations in comparison with P . vivax worldwide [9 , 10] . In this genome-wide analysis of nucleotide diversity π , we confirm that P . vivax-like populations are significantly more diverse than P . vivax populations ( P < 0 . 001; Wilcoxon test ) , with P . vivax-like samples showing nearly 10 times higher nucleotide diversity ( πP . vivax = 0 . 0012; πP . vivax-like = 0 . 0096 ) . Such genetic diversity in P . vivax-like strains in comparison with human P . vivax has already been described in other studies [9 , 10] , suggesting that P . vivax-like strains probably display extremely high genetic diversity in Central African regions . This suggests that P . vivax-like parasites of African great apes are probably more ancient than the human P . vivax strains and that the human P . vivax species ( as for human Plasmodium species like P . falciparum ) went through a bottleneck and only recently underwent population expansion .
Before the discovery of the ape P . vivax-like , the main hypothetical scenario concerning the origin of P . vivax was that of an “Out of Asia . ” More specifically , it was considered that P . vivax emerged in humans following its transfer to humans from Asian monkeys , as has been recently described for P . knowlesi [42] . However , the recent discovery of P . vivax-like in African great apes [10] and the analysis of their genetic characteristics led researchers to propose an “Out of Africa” origin of P . vivax [10 , 36] . Based on phylogenetic analyses of partial mitochondrial genomes and nuclear sequences of P . vivax and P . vivax-like parasites isolated from great apes and humans , Liu and colleagues [10] suggested that all extant human P . vivax parasites derived from one single ancestor that was transferred from great apes to humans . Our results do not bring any new evidence in favor of one or the other scenario . We think , nevertheless , that the origin of P . vivax is more complex than recently proposed ( a single host switch from apes to humans in Africa ) and that some previous data—especially those regarding the paraphyly of P . vivax-like and the inclusion of P . vivax within the diversity of P . vivax-like ( the arguments used in support of an African scenario ) —have been overinterpreted . The inclusion of P . vivax into the P . vivax-like diversity based on only a couple of nuclear genes and partial mitochondrial genomes is not a definitive proof of transfer of P . vivax-like to humans in Africa because alternative explanations can be provided . Indeed , as discussed above , such a phylogenetic pattern can be obtained because of ILS or because of a lack of phylogenetic signal for the sequences used , 2 phenomena that are frequent when species diverged recently ( which is the case for P . vivax and P . vivax-like ) . Our study indeed shows that incongruent phylogenies may be obtained when one limits the analyses to a single or a couple of genes ( see section above and S6 and S7 Figs ) . In our mind , it is still impossible to conclude that P . vivax has an African origin and that it was transferred from African apes to humans , especially when several observations are more in favor of an Asian origin , such as the following: ( i ) P . vivax evolved in a clade of parasites infecting Asian monkeys , and ( ii ) the highest genetic diversity of P . vivax is observed in Asian populations , and its diversity decreases toward Africa . This pattern of diversity is the opposite for P . falciparum , which has a well-established African origin [43] . For this parasite , the highest genetic diversity is found in Africa and decreases toward Asia accompanying the human migration [43] . Because the data that are currently available are not easy to interpret and are somehow contradictory , we think that more complex scenarios regarding the origin of P . vivax should be envisaged in light of phylogenetic and population genetic evidences ( as proposed in Prugnolle and colleagues [9] ) and more genomic data need to be obtained in support of these scenarios .
Comparison of the P . vivax genome to its closest sister lineage ( P . vivax-like ) and to the other primate Plasmodium parasites provides a unique opportunity to identify P . vivax–specific adaptations to humans . We applied a branch-site test of positive selection to detect events of positive selection that exclusively occurred in the P . vivax lineage . Within the reference genome P . vivax-like ( Pvl01 ) , 418 genes exhibited significant signals of positive selection ( S5 Table ) . In the human P . vivax genome PvP01 , the test allowed the identification of 255 genes showing significant signals of positive selection ( S6 Table ) . Among these genes presenting a significant dN/dS ratio , 71 were shared between P . vivax and P . vivax-like ( genes indicated in orange in the S6 Table ) , including 56 encoding for proteins with unknown function and 15 encoding for proteins that are involved either in energy metabolism regulation ( n = 9 ) , chromatid segregation ( n = 2 ) , or cellular-based movement ( n = 4 ) . We then took into consideration those 255 genes detected under positive selection in P . vivax and compared them to those obtained in P . falciparum ( 172 genes under selection; see Supplementary Table 4 in [13] ) . We identified a subset of 10 genes under positive selection in both the human P . vivax and P . falciparum parasites ( P < 0 . 05 ) ( S7 Table ) . Among these 10 genes , 5 code for conserved Plasmodium proteins with unknown function and 3 for proteins involved in either transcription or transduction . Interestingly , the 2 remaining genes under positive selection in these 2 human Plasmodium parasites code for the oocyst capsule protein , which is essential for malaria parasite survival in the Anopheles’ midgut , and for the rhoptry protein ROP14 , involved in protein maturation and the host cell invasion . These results suggest that these proteins could be essential for infection of humans or their vectors , and future studies should focus on the involvement of these proteins in human parasite transmission and infection .
Through technical accomplishments , we produced and assembled the first P . vivax-like reference genomes , the closest sister clade to human P . vivax—an indispensable step for a better understanding of this enigmatic species . We established that P . vivax-like parasites form a genetically distinct clade from P . vivax . Concerning the relative divergence dating , we estimated that the divergence between both species occurred probably before the split between P . malariae species . This suggests that the transfer of Plasmodium parasites to humans happened several times independently over the history of the Homo genus . Our genome-wide analyses provided new insights into the adaptive evolution of P . vivax . Indeed , we identified several key genes that exhibit signatures of positive selection exclusively in the human P . vivax parasites and show that some gene families important for RBC invasion have undergone species-specific evolution in the human parasite , e . g . , rbp and dbp . Are these genes the keys to the emergence of P . vivax in the human populations ? This pending question will need to be answered through functional studies associated with deeper whole-genome analyses . Among the genes identified to be under positive selection , 2 were also identified to be under positive selection in the other main human malaria agent , P . falciparum , thus suggesting their key role in the evolution of the parasites in their ability to infect humans or their anthropophilic vectors . To conclude , this study provides the foundation for further investigation into P . vivax traits that are of public health importance , such as features involved in host–parasite interactions , host specificity , and species-specific adaptations .
P . vivax-like samples were identified by molecular diagnostic testing during a continuous survey of great ape Plasmodium infections carried out in the Park of La Lékédi , in Gabon , by the Centre International de Recherches Médicales de Franceville ( CIRMF ) [9] . In parallel , a survey of Anopheles mosquitoes circulating in the same area was conducted in order to identify potential vectors of ape Plasmodium [11] . Specifically , mosquitoes were trapped with CDC light traps in the forest of the Park of La Lékédi . Anopheles specimens were retrieved and identified using a taxonomic key [44] before proceeding to dissection to isolate the abdomen . Samples were then stored at −20 °C until transportation to the CIRMF , Gabon , where they were stored at −80 °C until processed . Blood samples of great apes were treated using leukocyte depletion by CF11 cellulose column filtration [45] . P . vivax-like samples were identified either by amplifying and sequencing the Plasmodium Cytochrome b ( Cytb ) gene as described in Ollomo and colleagues [46] or directly from samples already used for studies of other Plasmodium species [13 , 14] . This allowed the detection of 11 P . vivax-like samples , 10 from chimpanzees and 1 from an An . moucheti mosquito . Most of these samples were co-infected with other Plasmodium species and/or probably with multiple P . vivax-like isolates ( see below and S1 Table ) . The identification of intraspecific P . vivax-like co-infections was made by analyzing the distribution of the RAF [47] . The animals’ well-being was guaranteed by the veterinarians of the “Parc of La Lékédi” and the CIRMF , who were responsible for the proceeding sanitary procedures ( including blood collection ) . All animal work was indeed conducted according to relevant national and international guidelines . These investigations were approved by the Government of the Republic of Gabon and by the Animal Life Administration of Libreville , Gabon ( no . CITES 00956 ) . It should be noted that our study did not involve randomization or blinding . To overcome host DNA contamination , within 6 hours after blood collection in the La Lékédi park , each 5 mL chimpanzee blood sample was diluted with 1 volume of PBS and passed through CF11 cellulose powder columns to remove host leukocytes ( i . e . , leukocyte depletion ) [45] . DNA was then extracted for the 11 samples using Qiagen Midi extraction kits ( Qiagen ) following the manufacturer’s recommendations . Next , DNA samples were enriched using a WGA using REPLI-g Mini Kit ( Qiagen ) following the strategy described in Oyola and colleagues [48] to optimize WGA of AT-rich genomes from low DNA quantities . The Illumina isolates were sequenced using Illumina Standard libraries of 200- to 300-bp fragments , and amplification-free libraries of 400- to 600-bp fragments were prepared and sequenced on the Illumina HiSeq 2500 and the MiSeq version 2 according to the manufacturer’s standard protocol ( S1 Table ) . Because of its high DNA content , the Pvl06 isolate was sequenced using PacBio chemistry . After a greater than 8-kb–size selection of DNA fragments using the BluePippin Size-Selection System ( Sage Science ) , the library for this Pvl06 sample was sequenced using P6 polymerase and PacBio chemistry version C3/P5 in 20 SMRT cells . Raw sequence data are deposited in the European Nucleotide Archive . The accession numbers can be found in S1 Table . For the P . vivax-like Pvl01 and Pvl06 , P . vivax PvP01 and SalI , P . cynomolgi B strain , and P . knowlesi H strain genomes , gene variants were detected and counted using Geneious software [60] . This allowed the presence/absence of the variants of the different gene families across Plasmodium species to be evaluated . For each gene family , the number of variants identified in the 2 reference genomes Pvl01 and Pvl06 was confirmed by manually checking the presence/absence of these in the other P . vivax-like genotypes obtained using ACT and bamview [61 , 62] . Orthologous groups across ( 1 ) P . vivax PvP01 [15] , P . vivax-like Pvl01 , P . cynomolgi B strain [8] , and P . knowlesi H strain [18] reference genomes and ( 2 ) the 13 Plasmodium reference genomes used for the phylogeny ( the here-generated P . vivax-like Pvl01 , P . vivax PvP01 [15] , P . cynomolgi M version 2 [17] , P . coatneyi strain Hackeri [63] , P . knowlesi H strain [18] , P . falciparum 3D7 [64] , P . praefalciparum G01 [13] , P . reichenowi PrCDC [65] , P . gallinaceum 8A [30] , P . ovale wallikeri PowCR01 , P . ovale curtisi PocCR01 , P . malariae PmUG01 , and P . malariae-like PmlGA01 [14] ) were identified using OrthoMCL version 2 . 09 [66 , 67] after an all-against-all BLASTp ( E-value threshold: 1 × 10−6 ) . From these , we extracted different sets of one-to-one orthologous genes for the subsequent analyses: a set of 4 , 056 genes that included the one-to-one orthologues among the 4 restricted species—P . vivax , P . vivax-like , P . cynomolgi , and P . knowlesi ( the Pv4sp set ) —and a set of 2 , 943 among the 13 Plasmodium species considered here for the interspecies phylogenetic analysis ( the Pl13sp set ) . The first set of orthologous groups identified was used for the detection of selection ( see below ) ; the second one was used to build the phylogeny and for the dating analyses . Amino acid sequences of the one-to-one orthologues were aligned using MUSCLE [68] or MAFFT [69] , respectively , for the first and second set of orthologous groups . Prior to aligning codon sequences , we removed the low-complexity regions identified on the nucleotide level using dustmasker [70] and then in amino acid sequences using segmasker [71] from ncbi-blast , using default parameters . After MUSCLE/MAFFT alignments , we finally excluded poorly aligned codon regions using Gblocks ( parameters: -t = p -b5 = h -p = n -b4 = 2 ) [72] . After going through all these alignment cleaning steps , we ended with a low number of missing data and no gap in our dataset of one-to-one orthologues ( 31 missing data over the 2 , 784 alignments of more than 50 amino acids ) . Most of the current methods available to estimate the timing of species splits make strong assumptions on the evolutionary models and often require an accurate mutation rate or calibration points , which are not always available . Here , we estimated the relative divergence times of Plasmodium species to free from temporal calibration of phylogenies and used 2 methods that do not depend on the estimation of a mutation rate: a method based on pairwise amino acid sequence divergences and Total Least Squares ( TLS ) regressions but assuming a constant rate of evolution across the Plasmodium lineages—the so-called Silva method [29]—and the RelTime method developed in Tamura et al . [31 , 32] , which does not assume any specific model for the rate of evolution . We first built the phylogenetic tree of the 13 Plasmodium species considered here for the dating analyses using the software RAxML version 8 . 2 . 8 [73] . The tree was constructed using the concatenation of the 2 , 784 orthologous groups that have a final alignment of more than 50 amino acids ( among the Pl13sp set of 2 , 943 groups ) . RAxML was called using the following options in order to automatically determine the substitution model that best fits the data model and so that 100 bootstraps were performed to assess the tree robustness: “-m PROTGAMMAAUTO -f a -# 100 . ” The best amino acid model of substitution identified by RAxML for the dataset ( the JTT model with empirical base frequencies JTT + F ) was used in the subsequent analyses . The idea behind the Silva method [29] for relative age estimation is that the divergence between amino acid sequences in independent lineages is correlated and that the divergence regression slopes of the proteins of a species pair of reference with the divergence of the proteins of other species pairs reflect the relative age of those splits ( see Silva and colleagues [29] for a detailed description of the method ) . Following this method , we obtained , for each of the 2 , 784 orthologous groups of the Pl13sp set that have a final alignment of more than 50 amino acids and for each species pair , the amino acid sequence divergence dAA through a pairwise comparison using PAML version 4 . 7 [74] , with the option “cleandata” set to 1 to remove sites with missing data from the dataset . The sequence divergences dAA were estimated using 4 different models of substitution ( JTT , WAG , LG , and Dayhoff ) to evaluate their influence on the estimates of the relative ages . An R script from the authors of this method [29] allowed the estimation of α , the slopes of the TLS regressions of the divergence of the proteins between every possible species pairs and the reference species pair , with the 95% CI by bootstrapping ( n = 10 , 000 bootstraps ) . The slope α is an estimator of the relative age of the 2 considered species , relative to the reference species pair . To evaluate the influence of the choice of the reference species pair on the results , we used different species pairs to consider multiple divergence references: We considered the relative distance of the split of the speciation of ( i ) P . vivax and P . knowlesi ( reference pair considered in the original paper ) , ( ii ) the 2 P . ovale species , and ( iii ) the P . malariae-like and P . malariae species . Because the model of Silva assumes a strict molecular clock [29]—which would not apply to all Plasmodium species , specifically P . falciparum because of its extreme GC content in comparison with other Plasmodium species—we used the clock calibration-free method RelTime [31 , 32] implemented in MEGA7 [75] to estimate divergence times when evolutionary rates vary . RelTime estimates relative rates for each branch of the tree without requiring a distribution of rate heterogeneity . The relative rates of 2 sister lineages are estimated assuming that the time to their most recent common ancestor is equal in the presence of contemporaneous sampling , and the rate of their ancestral branch is estimated as the average of their branch-specific relative rates . This framework is applied to subtrees of 3 or 4 taxa following a bottom-up strategy , from tips to the root to estimate relative rates and times , relative rates being scaled by setting the lineage relative rate of the ingroop root node to 1 ( see Tamura and colleagues [32] ) . RelTime analysis was performed on the concatenation of the alignments of the same 2 , 784 orthologous groups we used to build the phylogenetic tree , removing sites with missing data ( Complete-Deletion option ) and using the same substitution model ( JTT + F ) . As required by the RelTime analysis , we specified P . gallinaceum 8A [30] as the outgroup . The phylogenetic relationships of the RBPs in the P . cynomolgi–P . knowlesi–P . vivax–P . vivax-like group of species was reconstructed using a maximum likelihood analysis using RAxML [73] with 100 bootstrap replicates . The tree was visualized using Geneious software [60] . To investigate relationships between the P . vivax and P . vivax-like populations , we constructed a maximum likelihood tree using the filtered variant call set of SNVs limited to the higher allelic frequency genotypes identified within each sample using RAxML and PhyML ( using general time-reversible [GTR] models ) [73 , 76] . Trees were visualized using Geneious software [60] . All approaches showed the same final phylogenetic tree described in the “Relationships to worldwide human P . vivax isolates” section . We looked for signatures of introgression and/or ILS by performing a phylogenetic network using SplitsTree 4 . 14 . 6 [77] based on the alignment of a portion of the mitochondrial genome containing the cox1 and cytb genes , as in Liu and colleagues [10] . We included all but 2 ( KF618538 and KF618534 ) samples from Liu and colleagues [10] , all from Prugnolle and colleagues [9] , and 3 of our samples with a high-quality–sequenced mitochondrial genome ( Pvl08 , Pvl09 , and Pvl10 ) , and we added P . cynomolgi as the outgroup ( GenBank accession number: AB434919 . 1 ) . Gaps and monomorphic positions were excluded for the analysis , and the dataset resulted in an alignment of 85 sequences of 127 variable positions . We conducted a reticulate network using the RECOMB 2007 [78] and the NeighborNet distance transformation methods [79] . A phylogenetic tree based on the complete alignment , without exclusion of sites ( i . e . , an alignment of 2 , 530 bp ) , was constructed using FastTree 2 . 1 . 5 ( maximum likelihood ) implemented in Geneious [60] for comparison with results obtained from Liu and colleagues [10] . To explore the influence of the amount of phylogenetic signal on phylogenetic reconstruction , we also performed phylogenetic trees as described in the previous section using a subset of SNVs: Trees were reconstructed using 100 , 200 , 300 , 400 , 500 , 600 , 800 , 1 , 000 , or 5 , 000 SNVs . The nucleotide diversity ( π ) is the average number of nucleotide differences per site between 2 sequences . This parameter is interesting to estimate because it gives valuable information about variation in prevalence and demographic histories of the parasites . For the P . vivax and P . vivax-like populations , we calculated the genome-wide nucleotide diversity ( π ) from VCF files using VCFTools [59] . The means of nucleotide diversities was compared between P . vivax and P . vivax-like species based on the Wilcoxon-Mann-Whitney nonparametric test . In order to identify genomic regions involved in parasite adaptation to the human host , i . e . , regions under positive selection , we performed branch-site tests . To search for genes that have been subjected to positive selection in the P . vivax lineage alone , after the divergence from P . vivax-like , we used the updated branch-site test of positive selection [80] implemented in the package PAML version 4 . 4c [74] . This test detects sites that have undergone positive selection in a specific branch of the phylogenetic tree ( foreground branch ) using Likelihood Ratio Tests ( LRTs ) to compare models allowing positive selection or not . The selective pressure is defined as the ratio ( ω ) of nonsynonymous ( i . e . , dN , keeping the amino acid ) to synonymous ( i . e . , dS , changing the amino acid ) substitutions ( dN/dS ) . Under neutrality , rates of synonymous dS and nonsynonymous dN substitutions are equivalent , so the ω ratio is expected to be equal to 1 . Under purifying selection , the rate of nonsynonymous substitutions dN is expected to be lower than the rate of synonymous substitution dS because selection will prevent amino acid modifications: The ω ratio observed is then under 1 . This process is acting in reducing the fixation of deleterious mutation . Finally , under positive selection , amino acid replacement will be chosen , with the dN rate expected to be superior to the dS rate , leading to an observed ω ratio superior to 1 . This indicates the fixation of advantageous mutations . All coding sequences in the core genome with a one-to-one orthologous relationship among P . vivax Sal1 , P . vivax-like Pvl01 , P . knowlesi H strain , and P . cynomolgi B strain were used for this test ( i . e . , the 4 , 056 gene Pv4sp set of orthologous genes ) . We obtained dN/dS ratio estimates per branch and gene for P . vivax and P . vivax-like lineages alone using Codeml ( PAML version 4 . 4c [74] ) with a free-ratio model of evolution . Genes with a significant signal of positive selection in P . vivax only were compared to the ones obtained in P . falciparum from Otto and colleagues [13] ( S4 Table in Otto and colleagues [13] ) , in order to identify , e . g . , essential proteins for human or vector infection . The data are deposited in the Dryad repository ( doi:10 . 5061/dryad . 32tm1k4 ) [81] . | Among the 5 species responsible for malaria in humans , Plasmodium vivax is the most prevalent outside Africa . It causes severe and incapacitating clinical symptoms with significant effects on human health . Yet little is known about its evolution , adaptation , and emergence in humans . The recent discovery in African great apes of its closest known relative—P . vivax-like—may help resolve the evolutionary history of P . vivax . This study aims to characterize the genome of P . vivax-like isolated from infections of the closest ape relative to humans to get a better understanding of the evolution of this parasite . A total of 11 P . vivax-like samples were obtained from infected chimpanzee blood samples and an infected mosquito collected in Gabon . We generated , and here present , the first 2 new genomes of P . vivax-like and 9 additional draft sequences . Genome-wide analyses provide new insights into the biology and adaptive evolution of P . vivax to different host species . Indeed , they highlight lineage-specific evolution of some gene families involved in key steps of the life cycle of P . vivax . Analyses also revealed that the divergence between P . vivax and P . vivax-like occurred before the one between P . falciparum and its sister species P . praefalciparum . | [
"Abstract",
"Genome",
"assemblies",
"Gene",
"synteny",
"and",
"gene",
"composition",
"Phylogenetic",
"relationships",
"to",
"other",
"Relationships",
"to",
"worldwide",
"human",
"A",
"still",
"uncertain",
"origin",
"of",
"Conclusion",
"Materials",
"and",
"methods"
] | [
"taxonomy",
"parasite",
"groups",
"medicine",
"and",
"health",
"sciences",
"methods",
"and",
"resources",
"plasmodium",
"parasite",
"evolution",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"parasitology",
"apicomplexa",
"phylogenetics",
"data",
"management",
"protozoans",
"phylogenetic",
"analysis",
"mammalian",
"genomics",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"computer",
"and",
"information",
"sciences",
"sequence",
"alignment",
"bioinformatics",
"malarial",
"parasites",
"evolutionary",
"systematics",
"animal",
"genomics",
"eukaryota",
"database",
"and",
"informatics",
"methods",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"evolutionary",
"biology",
"organisms"
] | 2018 | Plasmodium vivax-like genome sequences shed new insights into Plasmodium vivax biology and evolution |
The agents of paracoccidioidomycosis , historically identified as Paracoccidioides brasiliensis , are in fact different phylogenetic species . This study aims to evaluate associations between Paracoccidioides phylogenetic species and corresponding clinical data . Paracoccidioides strains from INI/Fiocruz patients ( 1998–2016 ) were recovered . Socio-demographic , epidemiological , clinical , serological , therapeutic and prognostic data of the patients were collected to evaluate possible associations of these variables with the fungal species identified through partial sequencing of the ADP-ribosylation factor ( arf ) and the 43-kDa-glycoprotein ( gp43 ) genes . Fifty-four fungal strains were recovered from 47 patients , most ( 72 . 3% ) infected in Rio de Janeiro state , Brazil . Forty-one cases were caused by Paracoccidioides brasiliensis and six by Paracoccidioides americana ( former PS2 ) . P . brasiliensis was responsible for severe lymph abdominal forms , whereas patients infected with P . americana presented a high rate of adrenal involvement . However , no statistically significant associations were found for all variables studied . P . americana presented 100% reactivity to immunodiffusion , even when tested against antigens from other species , while negative results were observed in 9 ( 20% ) cases caused by P . brasiliensis , despite being tested against a homologous antigen . P . brasiliensis and P . americana are sympatric and share similar clinical features and habitat , where they may compete for similar hosts .
Paracoccidioidomycosis ( PCM ) is a mycotic disease with a large spectrum of clinical presentations that affects both immunocompetent and immunocompromised patients from different biomes of Latin America [1] . The infection primarily affects the lungs and is acquired by the inhalation of Paracoccidioides sp . conidia or mycelial fragments that becomes aerosolized after the soil perturbation [2 , 3] . Once inside the host , the fungus differentiates into pathogenic multi-budding yeast-like cells and this dimorphic process may lead to pathogenesis [4] . Incidence in endemic areas is associated with deforestation , armadillo hunting , and agriculture practices [1] . Massive soil removal during construction was recently suggested as another risk factor for the acquisition of PCM when a localized epidemic was recently reported among people living near highway construction [5] . Natural infections in a wide variety of animals have been reported in the literature , but recurrent isolation of the fungus from armadillos ( Xenarthra superorder . ) has led many authors to suggest that these mammals play an important role in the fungal life cycle and dispersion [6] . The infection can be controlled and/or cleared by the innate immune system after inhalation , and is thought to be the main reason why the majority of infections are asymptomatic [7] . However , it may progress to an acute/sub-acute disseminated pathology that may affect lymph nodes , liver , spleen , gut , bones joints , and meninges or to a chronic pulmonary disease [7] . The acute/sub-acute form may have several complications and sequelae such as low adrenal reserve , lymphedema , spleen calcifications , among others [8] . Moreover , the chronic pulmonary form of the disease can be disabling due to pulmonary fibrosis development , and is frequently seen in PCM endemic areas of Brazil , Argentina , Venezuela , and Colombia that report the majority of cases [9] . The number of early deaths , even in patients with unknown previous clinical history , is remarkably high and this mycotic disease is the most common cause of hospitalization due to fungal infections in immunocompetent patients in Brazil [10] . The severity of specific cases , the broad spectrum of clinical manifestations , and the highly variable immune response observed in patients with PCM requires further investigation of the organism’s genetic contribution to disease plasticity , diagnostics , and prognostics [11] . The etiological agents of human PCM are distributed into at least 5 species: Paracoccidioides brasiliensis ( former S1 phylogenetic group ) , Paracoccidioides americana ( former PS2 phylogenetic group ) , Paracoccidioides restrepiensis ( former PS3 phylogenetic group ) , Paracoccidioides venezuelensis ( former PS4 phylogenetic group ) , and Paracoccidioides lutzii ( former Pb01-like phylogenetic group ) [11–17] . The species P . restrepiensis and P . venezuelensis are geographically restricted to Colombia and Venezuela respectively , while P . brasiliensis , P . americana , and P . lutzii have a broad occurrence in Latin American countries [14 , 17] . Moreover , P . brasiliensis is composed of two cryptic populations: S1a and S1b , that are differentially prevalent along eastern Brazil and southern South America respectively [16 , 17] . P . americana has been identified in eastern Brazil and in a single occurrence in Venezuela , but with a lower incidence compared to P . brasiliensis [12 , 14 , 15] . P . lutzii constitutes a single genotype that is endemic to the north-mid western part of Brazil and Ecuador and it is genetically distant from the former P . brasiliensis species complex that includes the above mentioned species [11 , 13 , 15] . To date , fewer than 200 Paracoccidioides spp . strains have being properly genotyped and the availability of data from specific endemic areas of South America are scarce [12 , 13 , 14] . Regional efforts to understand the genetic epidemiology of these pathogens are needed as disease variations among patients are evident , treatment outcome may differ for different fungal species , and there may be differences in the arsenal of virulence factors expressed during infections within and between species . Approximately 80% of PCM cases are reported in the Brazilian territory [1 , 10] . The Southeast region of Brazil includes the states of São Paulo , Rio de Janeiro , Espírito Santo , and Minas Gerais , which are historically important areas of high endemicity of the disease [1] . The state of Rio de Janeiro has the third highest number of hospitalizations due to PCM in Brazil [10] . The disease was first reported by Adolpho Lutz in 1908 , and reporting over almost a century in this state reveals a strong association with rural lifestyle and farming [18 , 19] . The southern Paraíba Valley and the Resende basin are areas of sugar cane , coffee plantations and deforestation for both agriculture and livestock production , and are related to the disease burden in this state [20] . Acute PCM cases are also frequently reported in Rio de Janeiro suggesting active and constant dispersion of the fungus among the population [5 , 8] . The disease is highly endemic in the metropolitan area of the Rio de Janeiro state ( municipalities of Rio de Janeiro , Duque de Caxias , Itaguaí , Magé , Cachoeiras de Macacu , Nova Iguaçu ) as well as in the Paraiba Valley ( Volta Redonda and Barra Mansa ) [5 , 8 , 10] . However , hospitalizations due to PCM are recorded in the entire range of the state , even in mountain regions ( Petrópolis and Teresópolis ) , mid-South ( Vassouras and Paraíba do Sul ) or North of the state ( Campos dos Goytacazes and São Fidélis ) suggesting that the fungus is endemic to the whole state or perhaps the patients may have migrated from the place of infection to other regions of the state . Currently , 14 strains from patients living at Rio de Janeiro have been genotyped and the clinical information reported , resulting in 13 identified as P . brasiliensis and one as P . americana [8 , 21–25] . These species are potentially sympatric since they seem to inhabit the same geographical area , i . e . Rio de Janeiro and São Paulo states [8 , 14 , 22] . In order to investigate the genetic background of Paracoccidioides spp . and its possible medical associations in the endemic area of Rio de Janeiro , 54 clinical strains recovered from 47 patients presenting both chronic and acute forms of PCM were identified by molecular techniques and the epidemiologic , clinical , therapeutic , and serological features of the patients were associated with the identified species . We also compared the P . brasiliensis P . americana isolation ratio between São Paulo and Rio de Janeiro in order to better understand the ecology of those species complexes .
Isolation of fungi was carried out from 1998 to 2016 from patients admitted at the Evandro Chagas National Institute of Infectious Diseases ( INI/Fiocruz ) for diagnosis and clinical management of PCM . Paracoccidioides strains were isolated for diagnostic purposes from different human clinical samples: oral or nasal mucosa , lymph nodes , sputum , bronchoalveolar lavage , skin lesions , and spleen . The strains were maintained under mineral oil and recovered for molecular analyses after subcultures in Potato Dextrose Agar incubated at 25°C for 30 days . The Paracoccidioides spp . colonies were then subcultured in Fava-Netto agar plates incubated at 37°C for 14 days and single colonies were collected to further characterize the Paracoccidioides species by molecular biology . The use of anonymous patients’ data of the patients herein included was approved by the Ethics Committee Board of INI/Fiocruz ( number CAAE: 42590515 . 0 . 0000 . 5262 ) . All patients for whom at least one viable colony of Paracoccidioides sp . was cultured were included in this study . Medical records of the included patients were anonymously reviewed . Socio-demographic and epidemiologic data included age , sex , place of birth and residence of the patients , as well as the probable region where they became infected . The latter was inferred from the information provided by the patient such as place of birth , residence , development of risk activities such as agricultural and/or construction activities , or armadillo hunting . Clinical information included the PCM clinical form , the main organs affected , and the grade of disease’s severity according to Mendes et al . [26] . All patients underwent a standard routine clinical evaluation including physical examinations , blood tests [hematology , liver and renal function tests , Ouchterlony double immunodiffusion ( ID ) test for PCM , enzyme immunoassay tests for screening of HIV antibodies] , parasitological stool analysis , acid-fast bacilli and culture of clinical specimens , chest radiography , and other imaging examinations when indicated [brain computerized tomography ( CT ) , abdominal CT or ultrasonography] . The adrenal function was evaluated using the adrenocorticotropic hormone/ACTH stimulation test . Low adrenal reserve was defined as a cortisol normal basal level and levels lower than 20 mg/dl after 30 and 60 minutes of stimulation . Some of the above mentioned tests may not have been performed depending on test availability and patient’s consent . Therapeutic regimen was based on the Brazilian PCM guidelines [7] that were recently revised . The analyzed data included the drugs prescribed and the total time of treatment . Prognostic information was related to the patient’s outcome such as cure , relapses , complications , and death . Cure criteria considered clinical , radiological , and serological aspects [7] . Double-immunodiffusion ( ID ) assay was applied for specific antibody detection using a pool of crude antigens obtained from isolates Pb01 ( P . lutzii ) and Pb339 ( P . brasiliensis ) . The sera were obtained from the patients followed: at admission ( before treatment ) , every 3 months until cure was achieved , and every 6 months until patients’ discharge . A quantitative ID was performed , through 2-fold dilutions of sera in phosphate buffered saline solution . For comparison purposes , the serum titer of the first and last stored serum samples of each patient were compared . The genomic DNA of each strain was extracted from yeast-like cultures according to Ferrer et al . [27] and quantified using the NanoVue Plus Spectrophotometer . Each DNA sample was used as template for the PCR amplifications of the partial ADP-ribosylation factor ( arf ) and the 43-kDa-glycoprotein ( gp43 ) genes using the Platinum Taq DNA polymerase 2X PCR Master Mix . The mixture contained 5 μl of 10X reaction buffer solution , 1 μl of the forward ( ARF-F 5’TCTCATGGTTGGCCTCGATGCTGCC3’ and gp43-E2F 5’CCAGGAGGCGTGCAGGTGTCCC3’ ) and reverse ( ARF-R 5’GAGCCTCGACGACACGGTCACGATC3’ and gp43-E2R 5’GCCCCCTCCGTCTTCCATGTCC3’ ) primers ( 10 pM ) as described elsewhere [21] , 5 μl of deoxynucleoside triphosphate solution ( 0 . 2 mM ) , 2 μl of magnesium chloride solution ( 2 mM ) , 0 . 5 μl of Taq DNA polymerase ( 2 . 5 U ) , 100 ng of fungal genomic DNA , and ultrapure water in a final reaction volume of 50 μl . Annealing primer conditions and cycling were adapted from Teixeira and collaborators , as previously described [11 , 21] . The nucleotide sequences were determined via automatic capillary Sanger sequencing in an ABI 3730xl- Applied Biosystems machine using the BigDye Terminator v3 . 1 cycle sequencing kit ( Thermo Fisher Scientific , USA ) . Sequencing was performed using both forward and reverse primers [21] and nucleotide quality control was checked using Phred [28]; only called bases with a Phred score > 30 were considered for subsequent analysis . Representative sequences of arf and gp43 loci , covering P . brasiliensis ( S1a and S1b ) , P . americana , P . restrepiensis , P . venezuelensis , and P . lutzii were added to the dataset [11–13 , 29] ( S1 Table ) . The sequences were aligned using the ClustalW algorithm [30] implemented in the BioEdit software [31] and were manually inspected . In order to genetically classify the Paracoccidioides sp . strains from Rio de Janeiro state , Maximum Likelihood ( ML ) methods were applied . Phylogenetic trees were calculated using the IQ-TREE software [32] and nucleotide substitution models were selected using ModelFinder [33] . Each isolate was assigned to each species/genotype and branch support was inferred using both ultrafast bootstraps [34] and Shimodaira–Hasegawa approximate likelihood ratio test ( SH-aLRT ) . Trees as well branch supports were visualized using FigTree v1 . 4 . The haplotype networks were produced to visualize the microevolution of both P . brasiliensis and P . americana . The distribution and diversity of haplotypes for the arf + gp43 dataset was estimated using the software DnaSP , v 5 [35] and the Median-joining network was built and visualized used in Network , v 4 , software ( Fluxus Technology , Clare , Suffolk , England ) . Geographical locations and the genotypic profile of human , environmental and armadillo strains as well soil and biopsies amplicons recovered from P . americana and P . brasiliensis samples from Southeast Brazil ( Rio de Janeiro , São Paulo and Minas Gerais ) were retrieved for counts [12 , 14 , 36–42] . These three states are located in the Southeastern part of Brazil , a hot spot of PCM , and contain the majority of currently genotyped strains of Paracoccidioides . The sociodemographic , epidemiological , clinical , and prognostic data of the included patients were represented as frequencies and their respective 95% confidence intervals . If the mean values of a variable in two independent groups had confidence intervals that do not overlap , then the difference between the groups was considered significant . Additionally , Fisher exact test was used in the comparison of categorical data , Student’s t test in the comparison of treatment times , and the Wilcoxon test in the comparison of anti-Paracoccidioides antibody serum titers before and after treatment . The proportions of both P . brasiliensis and P . americana in each sample state of Brazil were compared using N-1 Pearson’s Chi-Square test with a 95% confidence interval [43] . P . brasiliensis and P . americana distributions were tested for deviation from the hypothetical ratio of 1:1 using a chi-square test in the Microsoft Excel platform . Finally , we tested deviation from the 1:1 ratio of the overall counts for humans and armadillo for host-specificity tests . The environmental records were excluded from the statistical analyses due to low sample size . In all analyses , a p<0 . 05 was considered to be statistically significant .
Laboratorial storage of Paracoccidioides remains a challenge , and the recovery rate of the fungus is usually low . A previous study has reported a 26% recovery rate in a collection of 70 P . brasiliensis strains maintained under mineral oil for long periods of time . Moreover , only strains stored less than 10 years were viable [44] . In the current study , from 128 clinical Paracoccidioides spp . strains stored under mineral oil in the mycological culture collection of INI/Fiocruz , 54 ( 42% ) remained viable after 1 to 17 years of storage and were recovered for molecular analysis , corresponding to a total of 47 patients . Phylogenetic analysis of 54 Paracoccidioides spp . clinical strains from Rio de Janeiro , Brazil indicates P . brasiliensis ( n = 48 ) and P . americana ( n = 6 ) as the causative agents of PCM in patients living in this state . Teixeira and collaborators [11] established that the partial sequencing of the genes arf and gp43 is able to differentiate P . lutzii from the three P . brasiliensis phylogenetic species ( S1 , PS2 , and PS3 ) . Years later , when these phylogenetic species were elevated to the formal taxonomic species P . brasiliensis , P . americana , and P . restrepiensis , respectively , it was shown that phylograms using nuclear concatenated coding loci , including arf and gp43 , can differentiate the newly described species [17] . In a clinical laboratory scenario , sequencing of several genes is a difficult task , and then the use of the two genes described in this work can facilitate Paracoccidioides species differentiation in a clinical setting . It is important to note that both arf and gp43 genes show a negative value for Tagima’s D , indicating an excess of low polymorphisms relative to expectation [17] . Moreover , our phylogenetic identification is supported by phylogenetic and haplotype network analyses that included arf and gp43 sequences derived from strains successfully identified in previous publications [11 , 12 , 17] , which support the concept that the sequencing of these two genes allows the differentiation between P . brasiliensis and P . americana . To the best of our knowledge , this is the largest assessment of P . americana ever reported and the main P . brasiliensis report of cases with molecular identification and description of the respective medical features of PCM . The geographic origin , that is , the probable place of infection , of P . brasiliensis strains included the following Brazilian states: Rio de Janeiro ( 34 cases ) , Minas Gerais ( 5 cases ) , Paraíba , Ceará , and Piauí ( 1 case , each ) . Cases of PCM due to P . americana were from Rio de Janeiro ( 5 cases ) and Minas Gerais ( 1 case ) . Fig 1 shows the geographic distribution of the Paracoccidioides spp . strains related to the probable source of infection of the cases here studied . The strains we define as belonging to the P . brasiliensis clade are genetically undifferentiated from the ones recovered previously from São Paulo state , Brazil . Also , the strains identified as P . americana , recovered from patients infected at the Rio de Janeiro state ( n = 5 ) or at Minas Gerais ( n = 1 ) , clustered in the same clade ( clade B ) with two other strains previously identified from São Paulo ( B7 and B23 ) . This clade is distinct from clade A , which contains strains mainly from São Paulo suggesting that cryptic genotypes within this species may exist within P . americana . Fig 2 represents the phylogenetic analysis of the strains identified in this study . The species identification based on haplotype network and phylogenetic analysis were similar ( Fig 3 ) . In brief , all P . americana strains from Rio de Janeiro and Minas Gerais clustered within a single haplotype complex ( Hap11 , Hap47 , Hap48 and Hap54 were collapsed ) within P . americana ( Fig 2 ) . The grouping of the strains B7 and B23 were also noted in the network analysis since those are also placed within the haplotype containing the newly P . americana defined strains . The P . americana type-strain isolated at São Paulo ( Pb3 ) grouped with a different haplotype . The majority of P . brasiliensis ( n = 46 , 95 . 8% ) strains genotyped were placed within the major haplotype complex of P . brasiliensis ( Hap1-2 , Hap10 , Hap41-43 , Hap45-46 , Hap49-53 and Hap55 were collapsed–Fig 2 ) . This included patients likely infected in Rio de Janeiro state as well as those that were likely infected in Paraíba and Ceará states . This large haplotype group also harbors the majority of clinical strains from São Paulo and Minas Gerais [12 , 14 , 40 , 41] , indicating that this is the predominant P . brasiliensis haplotype in Brazil . By analyzing the two loci , arf and gp43 , we did not differentiate the P . brasiliensis S1a and S1b cryptic genotypes as reported using whole genome typing [16] . One P . brasiliensis strain in this study ( 47735 ) formed an exclusive haplotype . This strain was isolated from a patient whose probable place of infection was the Rio de Janeiro state ( Duque de Caxias municipality ) , although the major clinical presentation of this case , portal hypertension , was very similar to another case caused by a strain belonging to the large P . brasiliensis haplotype [23] . A different P . brasiliensis strain in our study that did not cluster in this large haplotype ( 22027 ) was isolated from a patient living in Rio de Janeiro at the time of diagnosis , but who probably acquired the infection in the Brazilian Northeast state of Piauí , known for rare occurrences of autochthonous cases of PCM [1] . This patient presented the chronic form and did not report risk activities associated with PCM in the state of Rio de Janeiro or in any other state . However , in childhood , he lived in rural conditions for several years in his place of birth ( Teresina , the capital of Piauí state ) . This strain was genetically similar to the following reference strains: B11 ( an armadillo isolate , origin from Pará ) , B25 ( isolated from a chronic human case of São Paulo ) and U1 ( isolated from penguin feces , Antarctica ) and were also placed into a single branch on the phylogenetic analysis [12] . Considering the absence of PCM cases acquired in the northeastern semi-arid region [1] , we hypothesized that this patient may have traveled in transition areas of Brazilian savanna and Amazon biomes , in the northern region of Brazil , near his childhood home . The Southeastern region of Brazil is dominated by two main biomes: The Brazilian neotropical savanna and the Atlantic rainforest , with no clear geographic barriers that impair fungal migration through these areas . However , these two highland biomes are characterized by different climatic conditions , soil types , and a diverse floral and faunal composition . The three main highland Brazilian areas are: ( a ) The Atlantic Plateau , extending all along the eastern coast of Brazil , ( b ) Southern Plateau , advancing inland towards the southern and southern-central areas , and ( c ) the Central Plateau that is placed in the central regions of Brazil , which is mostly covered by the Brazilian savanna vegetation . For instance , past studies revealed that within the endemic area of Botucatu , São Paulo state , both humans and armadillos can be infected by both P . brasiliensis and P . americana [38] . A single armadillo captured in the surrounding areas of Botucatu , carried both P . brasiliensis and P . americana isolates suggesting that those species are likely sympatric . Both P . brasiliensis and P . americana were found in this study to be the causative agents of PCM in patients described in a recent outbreak of acute PCM occurring after the construction of a highway in the Rio de Janeiro metropolitan area [5] which reinforces that those species occupy the same geographical areas . In other to achieve a better understanding of the influence of ecological niches on speciation , more samples are needed , and a greater number of alleles need to be assessed , which may be facilitated by whole genome sequencing . Regarding the sociodemographic and epidemiological aspects , the mean age was 38 years of age ( 95% CI 34–42 ) for the cases due to P . brasiliensis , and 43 . 5 years ( CI 95% 33–54 ) among those due to P . americana . This is in accordance with several studies reporting that PCM affects mostly working adults [1 , 7 , 18] . There were no statistical differences between the variables analyzed related to the identified Paracoccidioides species . The main variables studied are detailed in Table 1 . No statistically significant differences were detected regarding clinical aspects and the species identified . Table 2 summarizes the main clinical findings of the cases studied according to the involved fungal strain . In Rio de Janeiro , the proportion of the acute form of PCM is historically reported as 3–10% of all PCM cases [5 , 8 , 18] . The high proportion of acute/subacute juvenile clinical forms as well as severe cases and HIV-AIDS coinfected patients in the present study are possibly related to the higher fungal burden of these cases that facilitates fungal isolation in culture , which was an inclusion criterion of this study . Both Paracoccidioides species identified in this study were involved in acute/subacute PCM cases . To the best of our knowledge , this is the first formal description of acute/subacute PCM due to P . americana . There is a previous report of a PCM case caused by P . brasiliensis PS2 ( now P . americana ) in a young male patient , however other clinical aspects of this infection were not reported , which impairs the correct classification of the clinical form in this case [14] . Due to the age of the patient , it is thought that this was an acute/subacute PCM case . Taken together , these two reports support the proposition that P . americana can also cause acute PCM . On average , adrenal involvement related to PCM is reported in 56% of autopsied cases [7 , 26] . Also , it is worth mentioning that the high proportion of adrenal impairment in the small group of cases due to P . americana suggests a possible adrenal tropism of this species . It is well described that P . brasiliensis [8 , 21 , 23 , 25] and P . lutzii [45] can be related to severe PCM cases . Adrenal involvement can bring severe sequelae , which also could associate P . americana as well as P . brasiliensis and P . lutzii with severe PCM cases . Table 3 shows the main characteristics of treatment and the respective total time of treatment according to each species . One P . brasiliensis infected patient did not receive antifungal treatment because she never returned to our institution after fungal isolation and diagnosis . The treatment times are in accordance with other studies on PCM therapy [7 , 26] . A previous study suggested that P . brasiliensis was less responsive than P . lutzii to SMZ/TMP using an in vitro susceptibility test [46] . In the present study the six P . brasiliensis infected patients had a good response to this drug . It is not possible to know the exact species within the former P . brasiliensis complex ( S1 , PS2 , PS3 , or PS4 ) studied by Hahn and collaborators , therefore this therapeutic response to SMZ/TMP needs to be further explored under the light of the new Paracoccidioides species . The high frequency of drug combination in this study reflects the severity and complexity of the cases since drug association is known to be a good strategy for critical and neurological cases [8] . Immunodiffusion ( ID ) was positive at the time of diagnosis in 80% of PCM cases caused by P . brasiliensis ( 95% CI 68–92 ) and in 100% of PCM cases caused by P . americana ( 95% CI 54–100 ) , showing no significant differences in sensitivity of ID regarding Paracoccidioides species . These results highlight the efficacy of ID in P . americana infected patients using antigens derived from other phylogenetic species . Fig 4 shows the reactivity of the test according to the etiologic agent in 38 patients who had paired titers for comparison . The first ( at admission ) and last ( at discharge ) available ID tests were considered , regardless of the outcome . For both groups , serum antibody titers at discharge were lower than at admission ( p values of <0 . 0001 and 0 . 0173 for P . brasiliensis and P . americana-infected patients , respectively ) . Eight cases caused by P . brasiliensis did not react to ID , although these results were expected for three cases due to HIV/aids coinfection [7] . Since aids can impair antibody production , reducing ID sensitivity , this parameter was also calculated excluding the 5 patients living with HIV/aids . In this scenario , ID of PCM caused by P . brasiliensis presented a sensitivity of 86% ( 95% CI 71–95 ) . Two previous reports regarding results of serological reactivity associated with the molecular species responsible for PCM reveal differences . While the former presents a case , whose fungal agent identified by serologic tools was P . lutzii [47] , the latter reports a case due to P . brasiliensis molecularly identified in which serology reacted only against P . lutzii antigens [48] . More studies are necessary to clarify these findings . However , our findings reinforce that molecular techniques are likely most appropriate . No differences between species were found regarding prognostic aspects ( Table 4 ) . Since its description , P . lutzii has been implicated in poor prognosis of PCM [15 , 45] . P . brasiliensis was responsible for many complications in the patients included in this study , severe cases including acute lymph abdominal forms and a fatal septic shock similar to a previous case report of PCM associated with P . lutzii [25 , 45] . P . americana was also associated to dysphonia and low adrenal reserve . In Botucatu , a municipality of São Paulo state , dysphonia was reported as a frequent PCM complication associated with laryngeal involvement [49] . It is not possible to infer the Paracoccidioides species associated with those cases , but Botucatu is located in the Brazilian southeast , an area of P . americana occurrence [7 , 22] . Another study conducted in São Paulo reported up to 44% significant hypoadrenalism in patients with PCM [50] , a frequency similar to that found in the P . americana patients of this study . In Rio de Janeiro , low adrenal reserve was observed in approximately 13% of patients with acute PCM [8] , a frequency similar to that observed in the P . brasiliensis infected patients herein described , which can be explained by the predomination of P . brasiliensis in the patients included in the present work . Microbial communities are made up of distinct genetic entities and defining species boundaries and range in fungal pathogens is essential for molecular epidemiology studies . Not only for its clinical relevance in Latin America , Paracoccidioides may offer an interesting model of complex genetic microbial entities . Strikingly , by analyzing our study and retrospective reports that used molecular techniques to differentiate the Paracoccidioides species , we observed that the ratio of P . brasiliensis/P . americana distribution is uneven considering either São Paulo , Rio de Janeiro , or Minas Gerais as distinct states as well taking account human or armadillo populations suggesting that those species may occupy different niches ( Fig 5 ) . According to our data , P . brasiliensis is more prevalent than P . americana in the three analyzed states of Brazil ranging from 2 . 6:1 in Minas Gerais , 7 . 5:1 in São Paulo and 8:1 in Rio de Janeiro ( Fig 5 ) . We compared those proportion via N-1 Chi-square test and observed that this uneven species distribution is statistically significant in São Paulo ( Difference—76 . 60% , 95% CI—42 . 7234 to 87 . 6818 , χ2–28 . 088 , DF 1 , P < 0 . 0001 ) and Rio de Janeiro ( Difference—78 . 80% , 95% CI—33 . 2949 to 90 . 0857 , χ2–17 . 656 , DF 1 , p < 0 . 0001 ) but not in Minas Gerais ( Difference—44 . 49% , 95% CI—-4 . 4054 to 71 . 7512 , χ2–2 . 81 , DF 1 , p = 0 . 0937 ) . We also used Chi-square tests in order look for deviations from the 1:1 ratio for the P . brasiliensis/P . americana species distribution . The uneven distribution was found significantly in São Paulo ( p < 0 . 0001 ) and Rio de Janeiro ( p < 0 . 0001 ) but not in Minas Gerais , where a tendency was noted ( p = 0 . 059 ) ( Fig 5 ) . We also observed a skewed species distribution by considering either humans ( 6 . 1:1 –p < 0 . 0001 ) or armadillos ( 18:1—p < 0 . 0001 ) , suggesting that in both mammal hosts this pattern is observed . Relevant differences in phenotypes have been already reported in the literature: ( i ) Strains from P . brasiliensis species produce more conidia compared to P . americana ( ii ) P . americana produces atypical yeast morphology at 37°C compared to P . brasiliensis species [14 , 17] . Variable loci may produce phenotypic plasticity in natural populations mainly due genetic drift . Under neutral selection , the genetic diversity inherited by a given population is dependent on the population size and mutation rate . In fungi , especially in dimorphic fungi , mutation rates and/or cell subdivision estimates are scarce making it difficult to determine effective population size . In sympatric species , such as P . brasiliensis and P . americana , understanding evolutionary aspects that may explain genetic plasticity is facilitated by comparing genomes of sister species that diverge in life history or ecology in the same geographical area . Recent population genomic studies revealed these closely related species have similar low nucleotide diversity indexes ( P . brasiliensis—π = 0 . 00053 and P . americana—π = 0 . 00066 ) and genome-wide calculation of Tajima’s D did not deviate from the null hypothesis suggesting neutrality . By combining this nucleotide diversity and phylogenomic measurements the authors suggested that P . americana is a more ancient species compared to P . brasiliensis [16] and this may impact the population size and fitness of this species . These observations , coupled with a skewed species distribution , suggest that P . brasiliensis may have more efficient mechanisms to survive in the environment and to infect mammals compared to P . americana . Moreover , those species may compete for the same host and thus it may explain the differences on population size of both species . In conclusion , 54 clinical strains were newly genotyped through sequencing of both arf and gp43 loci , reinforcing that both P . brasiliensis and P . americana are endemic species in Rio de Janeiro . The majority of P . brasiliensis strains from Rio de Janeiro state clustered within P . brasiliensis with no genetic differentiation from those from São Paulo . However , P . americana recovered from Rio de Janeiro formed a new cluster apart from that previously described containing strains from São Paulo , Minas Gerais ( Brazil ) , and Venezuela suggesting that the genetics of this species is more complex than previously thought . For the first time , clinical and molecular aspects of PCM in the endemic area of Rio de Janeiro are described . In this geographical region , P . brasiliensis was responsible for severe lymph abdominal forms including massive splenomegaly , portal hypertension and fatal septic shock . P . americana appears to have adrenal tropism , presented 100% reactivity to immunodiffusion , even when tested against antigens from other species , and caused acute forms , along with P . brasiliensis . No statistically significant associations were found between the two species analyzed and clinical aspects . Comparative analysis considering retrospective genotyped cases of human and armadillo infections suggests that those two species have different population sizes and may compete from the same host . This study was performed with regional-origin patients and present limitations with regards to the number of cases analyzed , especially in those caused by P . americana , as well as the representativeness of the species of the genus Paracoccidioides . However , the prevalence of P . americana in the Brazilian Services is unknown . Therefore , this number could not be so small , and the publication of these data may launch the knowledge of this species distribution and its clinical aspects . In addition , aspects such as the strain virulence , the inhaled fungal burden , the genetic and immunological susceptibility of the host as well as the diagnostic delay and other social determinants of health inequality deserve future studies to address their conjunct role in the spectrum of clinical presentations and in the severity of PCM . Future multicenter studies including a higher number of fungal strains including all species and their corresponding clinical data are required to fully understand this severe neglected systemic mycosis , so relevant to public health . | Paracoccidioidomycosis ( PCM ) is a severe systemic mycosis caused by different phylogenetic species . According to previous studies , these species could have an impact in PCM clinical features . This study aims to investigate possible associations between Paracoccidioides species and corresponding clinical data . The fungal strains from the patients were recovered , whereas their clinical data were collected to evaluate possible associations of these variables with the fungal species identified through DNA sequencing . Fifty-four fungal strains were recovered from 47 patients , most infected in Rio de Janeiro state , Brazil . Forty-one cases were caused by Paracoccidioides brasiliensis and six by Paracoccidioides americana . P . brasiliensis was responsible for severe clinical forms , and patients infected with P . americana presented a high rate of adrenal involvement . However , no statistically significant associations were found for all variables studied . P . brasiliensis and P . americana share similar clinical features . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"and",
"discussion"
] | [] | 2019 | Clinical features and genetic background of the sympatric species Paracoccidioides brasiliensis and Paracoccidioides americana |
As the Drosophila embryo transitions from the use of maternal RNAs to zygotic transcription , domains of open chromatin , with relatively low nucleosome density and specific histone marks , are established at promoters and enhancers involved in patterned embryonic transcription . However it remains unclear how regions of activity are established during early embryogenesis , and if they are the product of spatially restricted or ubiquitous processes . To shed light on this question , we probed chromatin accessibility across the anterior-posterior axis ( A-P ) of early Drosophila melanogaster embryos by applying a transposon based assay for chromatin accessibility ( ATAC-seq ) to anterior and posterior halves of hand-dissected , cellular blastoderm embryos . We find that genome-wide chromatin accessibility is highly similar between the two halves , with regions that manifest significant accessibility in one half of the embryo almost always accessible in the other half , even for promoters that are active in exclusively one half of the embryo . These data support previous studies that show that chromatin accessibility is not a direct result of activity , and point to a role for ubiquitous factors or processes in establishing chromatin accessibility at promoters in the early embryo . However , in concordance with similar works , we find that at enhancers active exclusively in one half of the embryo , we observe a significant skew towards greater accessibility in the region of their activity , highlighting the role of patterning factors such as Bicoid in this process .
During early embryogenesis all animal genomes undergo a transition from a largely quiescent to a highly active state with widespread zygotic transcription [1] . This process , known as the maternal-to-zygotic transition ( MZT ) , involves a major reorganization of chromatin , during which active and inactive regions are established which are distinguished by nucleosome composition , density and post-translational modifications [2–6] . It is generally thought that active—or “open”—chromatin facilitates the binding of polymerases , transcription factors and other proteins to target sequences , while inactive—or “closed”—chromatin limits the scope of their activity , although the degree to which chromatin state is instructive remains controversial [7 , 8] . Two important open questions are how genomic locations of active and inactive chromatin are encoded in the genome and how their active state is established , especially during the MZT , which follows early embryonic mitotic divisions where little or no differentiation into open and closed chromatin has been observed [2] . In Drosophila melanogaster , zygotic transcription largely begins at the seventh syncytial mitotic cycle ( although there is evidence for low levels of transcription from the beginning of embryogenesis [9] ) and gradually increases until the end of mitotic cycle 13 , when the embryo has several thousand nuclei and widespread zygotic transcription is observed [10 , 11] . Many of the genes activated during the MZT produce mRNAs that have spatially restricted distributions . These patterns are established through the activity of transcriptional enhancers , cis-regulatory sequences that integrate activating and repressive inputs from well-characterized , patterning transcription factors to produce novel , increasingly precise transcriptional outputs [12–15] . It is widely assumed that the interactions among patterning factors and the DNA to which they bind plays a central role in determining which sequences will function as enhancers , and that their competition with nucleosomes and recruitment of chromatin remodeling factors establish chromatin accessibility at selected sites [16–19] . The anterior morphogen Bicoid , for example , has been shown to create open chromatin at a subset of its targets [19] in the early embryo . However , we and others have shown that a parallel system involving the ubiquitously expressed , maternally-deposited pioneer factor Zelda also plays a role in this process [2 , 20–24] . Zelda binds prior to the MZT to a large fraction of the enhancers and promoters that become active once widespread zygotic transcription begins [20 , 25] . Most MZT enhancers and promoters contain conserved Zelda binding sites that are highly predictive of both transcription factor activity and chromatin accessibility [20] . Furthermore , Zelda binding is associated with changes to chromatin , including nucleosome depletion and specific post-translational modifications of histones [2 , 20–24] . Although abundant genetic , genomic and biochemical data support the importance of Zelda in establishing enhancer and promoter activity , many aspects of Zelda activity remain unresolved . While embryos lacking Zelda show severe defects prior to gastrulation , patterned , enhancer-driven transcription is not eliminated in Zelda- embryos [22 , 25–28] . This could reflect the activity of additional pioneer factors [21] such as the ubiquitously expressed trithorax-like/GAGA Factor ( or GAF ) which plays an important role in establishing accessibility at promoters [21 , 29–31] and is likely associated with changes in the nucleo-cytoplasmic ratio [32 , 33] . Because Zelda and GAF are ubiquitously expressed , while patterning factors have spatially restricted expression , we reasoned that we could probe their relative contributions to the establishment of chromatin accessibility by measuring spatial patterns of chromatin accessibility in the early embryo . As a first step towards this end , here we compare chromatin accessibility in anterior and posterior regions of the D . melanogaster embryo .
To determine the extent to which chromatin accessibility is spatially patterned along the A-P axis in the early embryo , we manually separated anterior and posterior embryo halves and performed a modified ATAC-seq [34] protocol on each half separately . Briefly , we collected cellular blastoderm embryos ( mitotic cycle 14 , embryonic stage 5 ) , flash froze them in liquid nitrogen , and then sliced each embryo at approximately 50% egg length ( ascertained by eye ) with a chilled scalpel , separating anterior and posterior halves into separate pools ( Fig 1A ) . We isolated nuclei from 20 anterior halves ( in duplicate ) , 20 posterior halves ( in duplicate ) , 10 frozen unsliced embryos , and a mixed sample containing a subset of nuclei from anterior and posterior samples and applied the ATAC-seq “tagmentation” process to each sample . We sequenced the resulting libraries , mapped reads to the D . melanogaster genome and normalized the data using standard methods ( S1 Fig ) . ATAC-seq accessibility profiles generated from sliced and unsliced whole embryo samples correlated highly , demonstrating that the slicing process does not introduce any biases ( rp = 0 . 95 , S2C Fig ) . Both halves correlate with published DNaseI hypersensitivity data from similar embryo stages [35] , demonstrating that our embryo preparation protocol coupled with ATAC-seq can accurately map accessibility in the equivalent of 10 whole frozen embryos ( rp > 0 . 78 , S2A Fig ) . Biological replicates of anterior and posterior halves that were collected , sliced , and tagmented independently moderately correlate with each other ( anterior replicates rp = 0 . 88 , posterior replicates rp = 0 . 80 , S2B Fig ) . To call peaks using MACS2 , we first merged replicates to increase the total read number and decrease spurious peaks that arise from low coverage regions [32 , 36] . We then filtered our peaks for those that were found in both replicates ( methods ) . Genome-wide , chromatin accessibility in the anterior and posterior halves is remarkably similar ( Fig 1B; rp = 0 . 94 on data binned into 1kb windows and rp = 0 . 90 for all whole embryo peaks S3 Fig ) . Dramatic changes in chromatin accessibility have been observed between early ( stage 5 ) and later stage ( stage 14 ) Drosophila embryos [17 , 35] . Expectedly , A-P halves are more similar to each other than embryos from stage 5 and 14 ( rp = 0 . 66 , S4 Fig ) . The conservation of chromatin accessibility patterns between halves is detailed in Fig 2 , which shows the results of our ATAC-seq experiments near loci of three A-P patterning genes ( even-skipped , giant , and hunchback ) and one dorsoventral patterning gene ( dpp ) . Each of these A-P loci contains enhancers that are active exclusively in the anterior or posterior half of the embryo ( denoted by colored in Fig 2 ) . For some , the peaks are of similar heights in both halves , such as at eve stripe 2 ( anterior ATAC-seq signal / posterior ATAC-seq Signal—755/686 ) . However there are some examples where accessibility is clearly reduced in the inactive half , such as at eve stripe 1 ( 1083/336 ) , the gt anterior enhancers 23 ( 648/211 ) and -10 ( 513/175 ) ( Fig 2 , marked by asterisks ) . To get a more systematic view of the relationship between transcriptional activity and spatial patterns of chromatin accessibility , we used available genome annotation and published in situ hybridization experiments to systematically identify A-P and dorsal-ventral ( D-V ) ( as a control ) patterning enhancers whose transcriptional outputs are restricted to one half of the embryo [37–48] ( S1 File ) . We excluded enhancers and promoters of genes expressed only around 50% egg length because the precision of manual slicing is most likely variable . We also excluded enhancers that did not overlap peaks called in any of the anterior , posterior , or whole samples , leaving 85 A-P and D-V patterning enhancers . Patterning enhancers clearly trend towards greater accessibility in the embryo half where they are active ( Fig 3 ) . Normalized ATAC-seq signal at anterior enhancers ( anterior rp = 0 . 81 ) is less correlated than all 1kb regions genome-wide ( gray; rp = 0 . 94 ) or D-V patterning enhancers ( rp = 0 . 97 ) while posterior patterning enhancers ( rp = 0 . 99 ) correlate similar to the genome-wide measurements ( anterior n = 30 , orange; posterior n = 9 , blue; dorsal n = 16 , purple; ventral n = 27 , green; Fig 3A and 3B ) . From this , it is clear that at anterior patterning enhancers , chromatin accessibility is greater in the anterior half . We computed a measure of differential accessibility ( positional skew score ) for each A-P enhancer by dividing the difference in accessibility in the anterior and posterior half by total accessibility , such that positive scores denote loci that are more accessible in the anterior half , negative scores signify loci that are more accessible in the posterior half , and loci with a score of zero have no difference in accessibility . We found that only anterior enhancers have a significantly greater mean positional skew score than D-V enhancers ( pant < 0 . 006 verses dorsal; 5 . 57e-05 versus ventral ) or random genomic regions with similar total accessibility ( pant < 6 . 68e-08 , Fig 3C and S1 Table ) . Accessibility at almost all anterior enhancers is skewed towards the anterior while that of posterior enhancers is skewed towards the posterior ( Fig 3D ) . This pattern is in contrast to D-V patterning enhancers and promoters ( S5 Fig ) and A-P patterning promoters ( Fig 4D ) . Although individually only six anterior enhancers had significant skews relative to random regions , it is remarkable that almost all of these enhancers are skewed towards the active half regardless of the degree of skew . Similar trends were seen when positional skew scores calculated from replicates were examined ( S6 Fig ) as well as in accessibility profiles derived from single embryo halves ( S7 Fig ) . In order to understand if this phenomena extends beyond annotated A-P patterning enhancers , we evaluated whether whole peaks that have chromatin accessibility skewed towards the anterior or posterior show specific patterning activity in the embryo . We overlapped significantly skewed peaks with regions identified in a genome-wide screen for enhancer activity [45] . We then utilized available in situ hybridization experiments to evaluate whether these fragments have spatially patterned activity ( http://enhancers . starklab . org/ ) . Indeed nine out of twelve significantly skewed enhancers showed spatially patterned activity ( 8 anterior and 1 posterior , S8D Fig ) . We next examined the promoters of A-P patterning genes using expression data from sections of embryos cryosliced along the A-P axis to curate lists of A-P patterning gene promoters [28] . Similar to our enhancer set , we only included promoters that overlapped accessibility peaks called in whole , anterior , or posterior samples and that are also associated with patterned expression confirmed by in situ hybridization assays ( n = 19 anterior promoters , n = 25 posterior promoters , S1 File ) . Though accessibility in the active and inactive halves is only slightly more similar at anterior promoters ( rp = 0 . 86 ) than at anterior enhancers ( rp = 0 . 81 ) , it is comparable to posterior promoters ( rp = 0 . 84 ) and D-V enhancers and promoters ( Fig 4A and 4B , dorsal promoters , rp = 0 . 93; ventral promoters , rp = 0 . 95 ) . We confirmed these trends by showing that the mean positional skew score of anterior and posterior patterning promoters is not significantly different than D-V patterning promoters or random regions ( Fig 4C and S1 Table ) . Though accessibility at promoters of A-P expressed , zygotic genes seems to show a very slight trend in the direction of activity , their means are not significantly different than that of random regions ( Fig 4C ) . What is more telling is that there is no distinct skew of accessibility in the expected direction of activity , in contrast to what we observed for A-P enhancers ( Fig 4D ) . From this we conclude that accessibility at promoters is not as correlated with transcriptional activity as enhancers . We then used published ChIP-seq data of A-P patterning factors from stage 5 Drosophila embryos [49] to examine binding patterns at similarly and differentially accessible A-P enhancers . We analyzed Bicoid , Caudal , Knirps , Giant , Hunchback , Kruppel , and Zelda binding data , normalized by the mean signal for each factor . A-P patterning enhancers that are more accessible in the anterior ( shades of orange ) generally are dominated by Bicoid binding , with strikingly little binding from other transcription factors , although there are some exceptions ( Fig 5 , S9 Fig ) . Enhancers more accessible in the posterior ( shades of blue ) generally have high Caudal , Knirps , Giant , and Kruppel binding , with more diversity in factors bound than at anteriorly accessible enhancers . Interestingly , enhancers with similar accessibility in both halves ( shades of white ) have a high diversity of factors binding—including Zelda ( Fig 5 , S9B Fig ) . These patterns reveal that while transcription factor binding clearly does not completely explain differential chromatin accessibility , there are clear differences in factor composition and density between differentially and similarly accessible enhancers . We next examined transcription factor binding at peaks called in whole embryo ATAC-seq samples . Using a stringent statistical cutoff , we found 107 anteriorly skewed peaks , 9 posteriorly skewed peaks , and 6640 peaks that were not significantly skewed . Anteriorly skewed peaks were enriched for Bicoid , GAF , CF2 and to a lesser extent Zelda binding sites while the unskewed peaks were enriched for Zelda , pnr ( GATA factor homolog ) , Dref ( associated with insulators ) , and CF2 . Due to so few peaks being skewed towards the posterior , none of the posterior peaks had significant motifs called , although GAF and Hunchback were enriched ( S8 Fig ) . We then overlapped transcription factor peaks with anteriorly skewed , posteriorly skewed , and unskewed peaks and found that there is a significant depletion in Hunchback , Kruppel , Caudal , Knirps , and Zelda peaks in the anteriorly skewed peaks ( p = 0 . 03 , 0 . 005 , 5 . 14E-10 , 0 . 01 , and 0 . 00001735 respectively , S8 Fig ) . These data further demonstrate the observation that , while transcription factor binding does not completely explain the differences between skewed and unskewed peaks , Bicoid , Zelda , and GAF are likely playing a role in shaping chromatin accessibility , as has already been reported by several recent studies [19 , 21 , 32] .
We designed this experiment to ask if the open chromatin observed at active enhancers and promoters is found in every nucleus , suggesting a dominant role for ubiquitous factors in establishing regions of genomic activity in the early embryo , or if open chromatin is spatially restricted , suggesting a dominant role for patterning factors . The data presented here offer several clarifying observations about the early Drosophila embryo . First , we find that the vast majority of regions observed to be accessible in whole embryos are equally accessible in anterior and posterior halves , including the promoters of many genes active in only one half . Second , for a curated set of enhancers driving patterns along the anterior-posterior axis , we find that chromatin is more accessible in nuclei where the enhancer is active . This is especially true of anterior patterning enhancers regulated by the primary anterior morphogen Bicoid , but is also observed for several posterior patterning enhancers . Third , in most cases where we see accessibility biased towards the embryo half where an element is active we also see clear chromatin in the “inactive” half . Each of these observations comes with the caveat that the signal measured in each pool of halves is an average from approximately 60 , 000 nuclei that clearly limits the conclusions that we can draw . For example , for a given , region equal accessibility in both halves could indicate uniform accessibility in nuclei across the embryo , but could also arise from similar fractions of nuclei active in both halves . Similarly , it is impossible to resolve whether quantitative differences between the halves are the result of different numbers of nuclei with open chromatin , differences in the levels of accessibility , or both . And , finally , we likely cannot detect cases where only a small fraction of nuclei are active , although with our current methods it is difficult to accurately estimate what our resolving power is . Nonetheless , a study published while this paper was under review that applied a single-nucleus chromatin accessibility assay , which is largely immune to these caveats , to a variety of embryonic stages largely confirms our finding [50] . They observe that early ( 2–4 hour ) chromatin is more homogeneous than at later embryonic stages , and that the majority of regions of open chromatin in blastoderm embryos show no clear spatial pattern . Furthermore , when they applied an unbiased clustering method to single nuclei ATAC-seq data , nuclei from blastoderm embryos separate into two broad populations , each with increased accessibility in enhancers with anterior and posterior biases respectively . We mapped their data to the regions we analyze above and find general agreement with ours ( Fig 6 ) . Skew scores computed with their data are correlated with our skew scores , although the skews from single-cell data are more extreme than those from the hand-dissected embryos ( Fig 6B ) . Notably , when these skew scores are plotted around A-P patterning regions , we detect a similar overall skew in the direction of activity at enhancers but not promoters ( Fig 6C and 6D ) . It is interesting that single cell data show greater magnitude of skew score at these regions . This likely reflects the increased spatial precision of single cell methods which are able to subdivide the embryo into anterior and posterior domains by accessibility profile instead of by approximately 50% egg length as we have done . Though there are numerous advantages to single-nucleus experiments , one of the benefits of the experiment reported here is that spatial information about pools of nuclei is determined independently from their accessibility profiles . As such , both experiments taken together reveal that though most genomic regions show similar patterns of accessibility , most A-P patterning enhancers are more accessible in their active half . The strong anterior skew we and others observe for Bicoid targets is consistent with a recent study showing that chromatin accessibility at a set of around 100 early embryonic enhancers is primarily dependent on Bicoid [19] . Given the strong anterior bias in Bicoid protein levels , it would have been surprising not to find an anterior bias in chromatin accessibility for these regions , although we note that there is significant chromatin accessibility in the posterior for many of these regions in both our data and that of [50] , perhaps reflecting activity by the low levels of Bicoid in the posterior [51] . A more comprehensive understanding of how chromatin accessibility is established requires better resolved data on how closely chromatin accessibility tracks with activity in these regions , as neither our data nor that of [50] currently provides greater spatial precision than anterior vs . posterior . There is a large body of literature showing , for example , that tissue specific enhancers often have open chromatin in tissues in which they are not active [52 , 53] , and that this is often a result of enhancer priming by pioneer factors . Furthermore , the transcriptional output of enhancer in the early Drosophila embryo [52] , and in many other systems , is determined by a balance between the binding of activators and repressors ( reviewed in [54] ) . We expect activators to be bound in parts of the embryo where the enhancer is active , but repressors will bind , by definition , in parts of the embryo where they are repressing enhancer activity . Assuming that open chromatin is associated both with activator and repressor binding implies that the nuclei where chromatin is open for a given enhancer will always be a superset of the nuclei where it is transcriptionally active—the question for the future is how wide these regions are and what their specific patterns tell us about the mechanisms of how they were established . It will also be interesting to look at the emergence of spatial patterns and biases over time . We have previously shown that nucleosome depletion at enhancers and other aspects of their chromatin state is established prior to the expression of most patterning factors [2] . It has also recently been shown that most enhancers and promoters for patterning genes are accessible by nuclear cycle 11 , that this state is maintained through DNA replication and mitosis , and that this open state is associated with ubiquitous factors Zelda and GAF [32] . This leaves open the possibility that the binding of ubiquitous pioneer factors plays a particularly important role in determining chromatin accessibility at early cycles . In conclusion , our data , as well as that of [19] and [50] , establish that though most genomic regions do not show any difference in accessibility , there is significant spatial patterning of chromatin along the anterior-posterior axis in blastoderm embryos , with a clear coupling of activity and accessibility at spatially patterned enhancers . Whether these patterns of chromatin accessibility are instructive for patterning transcription , or merely reflect patterns of activity , remains to be determined .
Drosophila melanogaster OregonR embryos were collected for 2 hours and aged for 90 minutes on molasses agar plates . Embryos were then dechorionated with 30%-50% bleach solution for three minutes . Embryos were hand staged at 20x magnification at 14°C to be mitotic cycle 14 ( NC14 ) using previously established methods [2] . NC14 embryos were placed in a custom freezing buffer consisting of ATAC-seq lysis buffer [34] without detergent , 5% glycerol , and 1ul of bromoblue dye . Embryos were then taken out of the freezing buffer and placed onto a glass slide which was then put on dry ice for 2–5 minutes . Once embryos were completely frozen , the glass slide was removed and embryos were sliced with a chilled razorblade . Sliced embryo halves were moved to tubes containing ATAC-seq lysis buffer with 0 . 15mM spermine added to help stabilize chromatin . Embryo halves were then homogenized using Kimble Kontes Pellet Pestle ( cat no . K749521-1590 ) . IGEPal CA-630 was added to a final concentration of 0 . 1% . After a 10 minute incubation , nuclei were spun down and resuspended in water . Twenty halves were added to the transposition reaction containing 25ul of 2x TD buffer ( Illumina ) , and 7 . 5ul of Tn5 enzyme ( Illumina ) . The reaction was incubated at 37°C for 30 minutes . Transposed DNA was purified using Qiagen Minelute kit . Libraries were then amplified using Phusion ( NEB cat no . F531S ) and Illumina Nextera index kit ( cat no . FC-121-1011 ) . Libraries were then purified with Ampure Beads at a 1 . 2: 1 beads to sample ratio and sequenced on the Hiseq4000 using 100bp paired end reads . Fragments over 500bp were removed from libraries using a Pippen prep to reduce sequencing bias with the Hiseq4000 . Fastq files were aligned to Drosophila dm3 genome with Bowtie2 using the following parameters -5 5–3 5 -N 1 -X 2000—local—very-sensitive-local . Mapping metrics are provided in supplementary S2 Table . Sam files were then sorted and converted to Bam files using Samtools , only keeping uniquely mapped reads with a MAPQ score of 30 or higher using -q 30 , proper pairs with -f 2 , and removing unmapped , not primary alignment , reads that fail platform vendor quality checks , and optical duplicates with sam flag -F 1804 . Duplicates were removed with Picard ( http://broadinstitute . github . io/picard/ ) . Bams were then converted to bed files with bedtools [55] and shifted using a custom shell script to reflect a 4bp increase on the plus strand and a 5bp decrease on the minus strand as recommended by [34] . Replicate bed files were merged . Finally shifted bed files were converted into wig files using custom scripts ( S3 File ) and wig files which were uploaded to the UCSC genome browser . Merged wig files were normalized to reflect 10 million mapped Drosophila melanogaster reads . Anterior and posterior samples were normalized by linear regression to the whole embryo sample not including the y intercept ( S1 Fig ) . A-P and D-V patterning enhancer and gene annotations were compiled from many sources ( S1 File ) [28 , 37–48 , 56] . In order to provide the most accurate promoter annotations possible for our analysis we used RACE , CAGE , and EST data performed in Drosophila melanogaster embryos [57] to identify promoters preferentially used by the fly embryo . When there were multiple promoters per gene ( as was frequently seen ) , we chose the promoter that was verified by all three methods , denoted by a “V” in Hoskins et . al . ( 2011 ) supplementary file 3 . There were three genes that did not have annotated promoters in the Hoskins et al . ( 2011 ) dataset that were used in our analysis . Instead , these promoter annotations came from the Eukaryotic Promoter Database converted to dm3 annotations [58 , 59] . In order to further validate our A-P and D-V patterning enhancer and promoter annotations we manually curated in situ hybridization images corresponding to 678 genomic regions from multiple sources [38 , 42 , 46–48 , 60–83] . Each region was manually inspected such that only regions with both an in situ hybridization image that shows spatially restricted expression as well as moderate accessibility signal ( wig signal > 200 ) were kept for further analysis leaving 253 enhancers and promoters with spatially restricted expression . Anterior and posterior patterning enhancers and promoters were categorized as either completely spatially restricted or mostly spatially restricted . A report PDF for each region , including in situ hybridization images , accessibility browser traces , and Z-score and p-value , are found in S2 File . Promoters used in our analysis were categorized as maternal , maternal-zygotic , or zygotic using previously published RNA-seq data from single embryos [11] . Only zygotically expressed gene promoters were used in Fig 4C . However , all classes of promoters were used in the rest of Fig 4 . All graphs were made with R scripts ( S3 File ) and Deeptools [84] . Accessibility skew score was calculated by the following equation: Xactive−XinactiveXactive+Xinactive ( 1 ) where Xactive is the wig signal in the half where the region is activating gene expression and Xinactive is wig signal in the half where the region is not supposed to activate gene expression . Accessibility skew score measures whether a region is differentially accessible in the expected direction . This score is useful when comparing differential accessibility regardless of which half is favored ( for example when comparing accessibility skew at anterior to posteriorly patterned regions ) . Positional skew score provides information about the direction of the skew such that regions that are more accessible in the anterior have a positive positional skew score while those that are more accessible in the posterior have a negative positional skew score . Positional skew score is calculated by the following equation: Xanterior−XposteriorXanterior+Xposterior ( 2 ) Where Xanterior is the wig signal in the anterior sample and Xposterior is the wig signal in the posterior sample . Significance for each region was determined by computationally matching each region to a random region that has the same total normalized wig score ( S10 Fig ) . Positional skew score was calculated for each random region ( termed RandSkewScore ) . These scores were distributed normally and were used to determine a Z-score for each region of interest ( ZROI ) by the following equation: zROI = AccessibilitySkewScore−μσ where μ is the mean of the random region distribution and σ is the standard deviation of the random region distribution . Two-tailed p-values were then calculated from the Z score . Replicates were merged and peaks were called on the merged bed file using MACS2 with the following parameters:—nomodel—call-summits—bdg -p 1e-3 . Reproducible peaks were selected using bedtools to intersect peaks called in both replicates and the merged samples . Anterior and posterior accessibility signal was averaged using custom scripts around the set of reproducible whole peaks and positional skew scores were calculated for each peak region as described above . Significantly skewed peaks were determined using random regions as with A-P patterning regions . Significantly skewed anterior and posterior peaks as well as unskewed peaks were intersected with experimentally derived promoters , Kvon predicted enhancers [45] , and transcription factor ChIP-seq binding data [15 , 49 , 85] . One-tailed fisher exact tests were performed to determine whether there was a significant depletion or enrichment of transcription factor peaks in significantly skewed accessibility peaks . Regulatory sequence analysis tools ( RSAT ) peak motif tool was used to find motifs in each peak set . Wig files from previously published ChIP-seq data were obtained for Kruppel , Hunchback , Giant , Knirps , Caudal , Bicoid [15 , 49 , 85] , and Zelda data from stage 3 , 4 , and 5 embryos [20] . Wig files were normalized by the mean signal for each sample , assuming that the mean signal over the entire genome is similar to that of background . This normalization essentially transforms the data into deviations from the mean such that signal from different experiments can be compared to each other . Wig signal around regions of interest was determined and graphed in R ( S3 File ) . For the heat maps , normalized wig signal was averaged over 3kb windows around regions of interest for each factor before being scaled such that the region with the highest value is equal to 1 and the lowest to 0 for each factor . For all DnaseI comparisons to ATAC-seq ( S2 Fig and Fig 5 ) , previously published DnaseI data from stage 5 embryos ( replicate 1 ) normalized to 10 million reads was used . For S4 Fig , DNaseI data was downloaded from the following SRA datasets: SRA:SRP002474 . 1 , SRA:SRX020691 . 4 , SRA:SRX020692 . 1 , SRA:SRX020693 . 1 , SRA:SRX020694 . 1 , SRA:SRX020695 . 1 , SRA:SRX020696 . 1 , SRA:SRX020697 . 1 , SRA:SRX020698 . 1 , SRA:SRX020699 . 1 , SRA:SRX020700 . 1 , SRA:SRX041410 . Reads were processed similarly to ATAC-seq data . Reads were aligned to the dm3 genome with Bowtie2 with the following parameters -p 10–5 5–3 5 -N 1 -X 2000—local—very-sensitive-local . Sam files were then sorted and converted to Bam files using Samtools using the same filters as ATAC-seq samples . Duplicates were removed with Picard . Bams were then converted to bed files using Bedtools and converted into wig files using custom scripts ( S1 Code ) . All replicates were merged and were normalized to 10 million mapped reads . Bam files labeled to correspond with published clusters were shared by [50] with the authors . Bam files corresponding to anterior and posterior clusters were merged respectively . Merged bam files were then converted into wig files and then normalized to 1 million reads . The anterior wig file was then normalized by linear regression to the posterior wig file . Positional skew score was measured for all A-P and D-V patterning regions in the same manner as was done for halves ATAC-seq data ( S4 File ) . | DNA in the nuclei of animals and other eukaryotes is not floating around freely . Rather it is wrapped around proteins called histones that are in turn compacted into higher order structures called chromatin . Highly compact chromatin can restrict which genes and regulatory regions are accessible to the machinery that turns DNA into RNA , effectively shutting genes off . In contrast , chromatin at active sequences tends to be more open and accessible . However , it is still unknown whether open chromatin enables genes to be active , or if open chromatin is a byproduct of activity . Here we use a genomic technique to compare chromatin in the anterior and posterior halves of young fly ( Drosophila melanogaster ) embryos and show that two important classes of regulatory sequences—promoters and enhancers—are accessible in parts of the embryo where they are transcriptionally active and , often , in parts where they are not . This suggests that chromatin accessibility is not a direct consequence of activity per se , and supports a model in which chromatin around active genes and enhancers is systematically opened by specific regulator proteins and is later refined by activity . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods",
"and",
"materials"
] | [
"invertebrates",
"in",
"situ",
"hybridization",
"molecular",
"probe",
"techniques",
"gene",
"regulation",
"regulatory",
"proteins",
"dna-binding",
"proteins",
"animals",
"invertebrate",
"genomics",
"animal",
"models",
"developmental",
"biology",
"drosophila",
"melanogaster",
"model",
"organisms",
"experimental",
"organism",
"systems",
"transcription",
"factors",
"epigenetics",
"molecular",
"biology",
"techniques",
"embryos",
"chromatin",
"drosophila",
"research",
"and",
"analysis",
"methods",
"embryology",
"genomic",
"signal",
"processing",
"probe",
"hybridization",
"chromosome",
"biology",
"proteins",
"gene",
"expression",
"molecular",
"biology",
"insects",
"animal",
"genomics",
"arthropoda",
"biochemistry",
"signal",
"transduction",
"eukaryota",
"cell",
"biology",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"cell",
"signaling",
"organisms"
] | 2018 | Patterns of chromatin accessibility along the anterior-posterior axis in the early Drosophila embryo |
Increasing energy expenditure through brown adipocyte recruitment is a promising approach to combat obesity . We report here the comprehensive profiling of the epigenome and transcriptome throughout the lineage commitment and differentiation of C3H10T1/2 mesenchymal stem cell line into brown adipocytes . Through direct comparison to datasets from differentiating white adipocytes , we systematically identify stage- and lineage-specific coding genes , lncRNAs and microRNAs . Utilizing chromatin state maps , we also define stage- and lineage-specific enhancers , including super-enhancers , and their associated transcription factor binding motifs and genes . Through these analyses , we found that in brown adipocytes , brown lineage-specific genes are pre-marked by both H3K4me1 and H3K27me3 , and the removal of H3K27me3 at the late stage is necessary but not sufficient to promote brown gene expression , while the pre-deposition of H3K4me1 plays an essential role in poising the brown genes for expression in mature brown cells . Moreover , we identify SOX13 as part of a p38 MAPK dependent transcriptional response mediating early brown cell lineage commitment . We also identify and subsequently validate PIM1 , SIX1 and RREB1 as novel regulators promoting brown adipogenesis . Finally , we show that SIX1 binds to adipogenic and brown marker genes and interacts with C/EBPα , C/EBPβ and EBF2 , suggesting their functional cooperation during adipogenesis .
Obesity and its associated metabolic complications such as diabetes are increasingly responsible for significant economic and social burdens in many countries worldwide . Physiologically , obesity develops when energy intake exceeds energy expenditure , and the current treatments of obesity have been primarily focused on reducing energy intake . Unfortunately , these measures were largely inefficient in maintaining long-term weight loss [1] . The recent discovery of thermogenic adipocytes [2–5] capable of burning fat in adult humans has provided an exciting new therapeutic approach for the treatment or prevention of obesity by increasing energy expenditure [6] . Fat cells are derived from multipotent mesenchymal stem cells ( MSCs ) , which can give rise to muscle , adipose , bone , or cartilage cells when given appropriate environmental cues . These cells can be broadly divided into fat storage cells , such as white adipocytes ( WA ) ; and fat burning cells , which include classical and inducible brown adipocytes ( BA ) ( also known as beige or brite adipocytes ) [7] . Brown cells contain high density of mitochondria and dissipate chemical energy as heat through the action of the mitochondrial protein UCP1 ( uncoupling protein 1 ) . It is evident that increased activity of thermogenic brown cells has beneficial effects on whole body metabolic homeostasis [3 , 5 , 8] and various environmental cues such as cold exposure and chemical activation of the β-adrenergic pathway can significantly up-regulate BAT activity [2 , 9] . Over the last couple of years , a number of protein factors as well as long non-coding RNAs ( lncRNAs ) and microRNAs have been identified as regulators in this process . For example , the members of the bone morphogenetic protein ( BMP ) family , the PPARγ co-factor PGC1α , the transcription factors ( TFs ) PRDM16 , EBF2 , KLF11 , the protein deacetylase SIRT1 , the secreted factors IRISIN and FGF21 as well as lncRNAs Blnc1 , lncBATE1 , and microRNAs miR193/365 have been shown to be essential for thermogenic fat cell recruitment [6 , 10–13] . To promote thermogenic adipocyte recruitment , it is necessary to have a fundamental understanding of the gene regulation networks that control brown and white adipogenesis and identify the key differences between these two morphologically similar but functionally distinct cell types . Gene regulation networks are composed of cis-regulatory elements and trans-regulatory protein factors . In response to environmental stimuli , trans-factors bind to cis-elements such as enhancers or silencers to modulate gene expression . Given the large size and complexity of mammalian genomes , it has been difficult to systematically identify cis-regulatory elements at the genome-wide level . The recent discovery of signature histone modifications for these cis-elements ( eg . H3K27 acetylation for active enhancers ) and the advance in massive parallel DNA sequencing facilitated to comprehensively define these elements . In addition to typical enhancers , a group of so-called super-enhancers was discovered recently . These super-enhancers are large enhancer clusters containing high-density transcription factor binding and are associated with cell type specific genes [14] . They are also stronger in terms of gene activation ability and play key roles in controlling cell identity in mammals . Trans-regulatory factors are mainly identified through differential gene expression and genetic analyses . For example , the prominent adipogenic factors PPARγ and C/EBPα are strongly up-regulated during the course of adipogenesis . Recently some of the TFs were also discovered by analyzing the TF binding motif in cell type specific enhancer elements . Through this approach , PLZF and SRF were identified as anti-adipogenic factors [15] while EBF2 was identified as an activator for brown adipogenesis [16] . Finally , the presence of a super-enhancer at the proximity of Klf11 gene in human adipocytes led to its identification as a browning factor [10] . Although previous studies have looked at individual histone modification ( H3K27ac ) occupancy in mature brite cells [10] and brown adipose tissue [17] , a comprehensive profiling including other important chromatin marks during brown adipogenesis is still lacking . For example , H3K4me3 marks the promoters of actively transcribed genes [18]; H3K4me1 and H3K9ac are found at active/poised enhancers/promoters [19]; In contrast , H3K27me3 is a repressive mark and enriches at polycomb-repressed loci [20] . This information is important for describing epigenomic landscapes and is often required for downstream bioinformatic analyses . For instance , active enhancer regions are defined by the presence of H3K27ac and the absence of H3K4me3 [21] , and bivalent domains are enriched by both H3K4me3 and H3K27me3 [22] . Extensive studies have been carried out to characterize the dynamic chromatin regulation of WA differentiation [15] , but the global chromatin landscape for BA lineage specification and differentiation is far from complete . Especially , the epigenomic transition state that is not observed in cells either at the beginning or end of adipogenesis and which is essential for the transmission of adipogenic signals [23] , has not been analyzed in detail in the brown lineage . In this study , we adapted a protocol to efficiently differentiate the murine MSC line C3H10T1/2 from multipotent precursors into mature BAs via pre-treatment with BMP7 . The secreted factor BMP7 has been shown to be essential for brown adipogenesis in vitro and in vivo and important for BA lineage specification [24] . We collected samples at five important time points , representing specific developmental stages ranging from ( 1 ) multipotent mesenchymal stem cells , ( 2 ) committed brown preadipocytes , cells during ( 3 ) early and ( 4 ) late intermediary transition states , to ( 5 ) mature BAs . Using these samples , we generated epigenomic maps for a number of key histone modifications , PPARγ binding , together with corresponding gene expression profiles including mRNAs , lncRNAs and microRNAs . To identify regulators specific for brown adipogenesis , we compared the BA dataset to the results from white adipogenesis and performed analysis of: ( 1 ) stage- and lineage-specific gene expression; ( 2 ) stage- and lineage-specific enhancer enriched TF binding motifs; ( 3 ) lineage-specific super-enhancers; ( 4 ) BMP7 responsive genes . Through these analyses we not only re-discovered most established regulators of brown adipogenesis , but also identified a number of novel putative activators of BA differentiation including the kinase PIM1 , and three TFs , SIX1 , RREB1 and SOX13 . Via gain- and loss-of-function analyses , we validated that these factors are indeed essential for BA lineage commitment or differentiation . Finally , we mapped and analyzed the genome-wide binding of SIX1 in BAs using ChIP-seq . We found that its binding sites are enriched for C/EBP and EBF motifs and co-immunoprecipitation ( co-IP ) experiments confirmed that SIX1 physically interacts with C/EBPα , C/EBPβ and EBF2 , suggesting their cooperation during BA differentiation . Moreover , through analysis of the chromatin dynamics at brown lineage-specific genes , we found these genes to be pre-marked by both H3K4me1 and H3K27me3 , and H3K27 demethylation at the late stage was not sufficient to promote their expression , indicating an essential role for H3K4me1 in poising brown genes for expression . In summary , we provide a comprehensive reference map for the dynamic epigenome and transcriptome during BA differentiation , propose a conceptual model of brown gene regulation , and also show that comparative transcriptomic and epigenomic analysis is a powerful tool for the discovery of novel regulators , resulting in the identification of four activators of BA differentiation in this study .
To examine the molecular control of cell fate transitioning from uncommitted progenitor cells to BAs , we used C3H10T1/2 MSCs and differentiated them into BAs , following a previously established protocol [24] , where the multipotent progenitors were first committed to the brown lineage by BMP7 treatment before differentiation was triggered using a chemical cocktail ( Fig 1A ) . In addition , the well-established WA differentiation model 3T3-L1 was included in this study for comparison of events during differentiation . We first confirmed the efficiency and specificity of our differentiation systems by visual inspection of cell morphology , by qRT-PCR , and Western blot analyses of lineage marker gene expression ( S1 Fig ) . The differentiation process for both lineages was highly efficient , as virtually all cells accumulated lipid droplets by day 7 of differentiation ( S1A and S1B Fig ) . As expected , Pparg2 , the master regulator of adipocyte differentiation , and other adipogenic marker genes such as Fabp4 , CD36 , Lpl , Adipoq and Cebpa were strongly up-regulated in both lineages in response to differentiation signals . In contrast , BA marker genes including Ucp1 , Cidea , Elovl3 , Ppara and Prdm16 were activated only in mature BAs , and the mitochondrial marker genes Cox7a1 and Cox8b were expressed much higher in mature BAs ( S1C and S1D Fig ) . Corresponding expression patterns were also detected at the protein levels for PPARγ , UCP1 , PPARα and CIDEA ( S1E Fig ) . These data indicated that our differentiation processes were specific and efficient . Next we profiled the transcriptome and epigenome during murine brown adipogenesis at five key time points: ( 1 ) day -3 ( d-3 , uncommitted progenitors ) ; ( 2 ) day 0 ( d0 , end of brown lineage commitment by BMP7 treatment ) ; ( 3 ) 6 hours ( 6h , end of epigenomic transition [23] ) ; ( 4 ) day 2 ( d2 , early BA differentiation ) and ( 5 ) day 7 ( d7 , mature BAs ) ( Fig 1A ) . For transcriptome profiling , we used RNA-seq for mRNAs and lncRNAs , and an array-based method for microRNAs . To validate our transcriptomic analysis during brown adipogenesis , a second replicate of the RNA-seq experiment was performed and the results indicated that the data were highly reproducible ( S1 Table ) . In parallel , we also profiled the transcriptome using RNA-seq for mRNAs and lncRNAs , and microarray for microRNAs during 3T3-L1 WA differentiation . When compared with the transcriptomes of mouse adipose tissues , we found that our in vitro BA and WA systems are closely related to their corresponding in vivo tissues ( S1 Table ) . To complement the analysis of transcriptional changes during brown adipogenesis , we also performed a comprehensive profiling of the dynamically changing chromatin landscape during BA differentiation by ChIP-seq . In this effort , we mapped a number of key chromatin marks including H3K4me1 , H3K4me3 , H3K9ac , H3K27ac and H3K27me3 during BA differentiation . We also performed replicates at two key time points ( d0 and d7 ) for all histone marks and the results showed that our ChIP-seq data were highly reproducible ( see S1 Table and S2A–S2C Fig ) . In addition , we profiled PPARγ binding using ChIP-seq in mature BAs where it is highly expressed . Examples of the epigenomic and transcriptomic landscapes as well as PPARγ binding during BA differentiation at the brown selective genes Cidea , Ucp1 and Ppara are shown in Fig 1B and S3 Fig . A corresponding epigenomic dataset for WA differentiation has been generated previously [15] and was used for subsequent comparative analyses . Prior to further analysis we validated our ChIP-seq datasets by examining the correlations between gene expression and various histone modifications . As expected , we found that highly transcribed genes were marked by active chromatin marks ( H3K9ac , H3K4me1 , H3K4me3 , and H3K27ac ) but not the repressive mark H3K27me3 at their promoters ( S1 Table and S2D Fig ) . Together , these datasets constituted comprehensive reference maps of the epigenome and transcriptome for both BA and WA differentiation . We first focused on genes that were dynamically regulated at different stages during adipogenesis , following the rationale that differentiation stage-specific expression mirrors functional roles for those genes . To this end , we systematically examined coding genes , lncRNA genes , and microRNA genes ( Fig 2 , S4 and S5 Figs , S2 Table ) . Using an entropy based method ( See “Materials and Methods” section for details ) , we identified a total of 2277 ( BA ) and 1513 ( WA ) differentiation stage-specific coding genes ( FPKM>5 ) , which were 26 . 2% and 16 . 0% of the expressed genes in the respective lineages . The higher number and proportion of genes with dynamic expression in BA are in agreement with the requirement of executing additional gene programs to commit MSCs into the adipogenic lineage before differentiation . We noted a clear separation into five stages with little overlap of stage-specifically expressed genes in WAs indicating a strictly step-wise differentiation process . Specifically , 3T3-L1 cells are at the proliferation stage at d-4; at d0 , these cells have been under growth arrest for 2 days [25]; after the adipogenic induction , the arrested cells re-enter the cell cycle and undergo an epigenomic transition stage at 6h [23]; while at d2 , these cells are arrested again and start to differentiate [25]; finally at d7 , 3T3-L1 cells are fully differentiated into mature adipocytes . In contrast , in BAs we observed a more substantial transition in gene expression between 6h and d2 , whereas the time points before ( d-3 , d0 , 6h ) , and after ( d2 , d7 ) showed a certain overlap of gene expression . This differentiation stage-specific gene expression pattern is likely derived from the fact that C3H10T1/2 cells continuously proliferate from d-3 to 6h without the contact inhibition and growth arrest stages observed in 3T3-L1 cells , and these cells start to accumulate fat earlier than 3T3-L1 cells ( S1A and S1B Fig ) at d2 . To analyze our observations more systematically , we performed gene ontology ( GO ) analysis of stage-specific genes . The top category of enriched genes before differentiation ( d-3/d-4 ) was “cell cycle” in both lineages , whereas the transient enrichment of “chondrocyte differentiation” was only found in BA after lineage commitment . Strikingly , the same enriched gene categories topped the list in BA at d2 and d7 , i . e . “brown fat cell differentiation” , “mitochondrion” , and “lipid metabolic process”; but in WAs “fat cell differentiation” tops the list not before d7 . This correlates with a well advanced differentiation status and accumulation of lipid droplets by d2 in BA , but not WA ( see S1A and S1B Fig ) , and may explain similar gene expression pattern between d2 and d7 in BA . We also examined lncRNA genes that are dynamically regulated during adipogenesis ( S4B Fig and S2 Table ) . Using the NONCODE database [26] , we found in total 1985 and 2796 expressed putative lncRNA genes during BA and WA differentiation ( FPKM>0 . 5 ) , respectively . Among them , 857 ( BA ) and 1135 ( WA ) lncRNAs showed a stage-specific expression pattern , which were 43 . 2% and 40 . 6% of the expressed lncRNAs , respectively . The proportions of stage-specific lncRNAs in BA and WA were therefore considerably higher than the ones for mRNAs which were 26 . 2% and 16 . 0% , suggesting lncRNA genes were regulated more dynamically during adipogenesis than coding genes . This trend was maintained even when the comparison was limited to lowly expressed mRNAs with similar expression levels as lncRNAs , which turned out to be the least dynamic between stages ( 15 . 5% and 10 . 8% ) . In addition , we also found a previously identified lncRNA ( Blnc1 ) that drives thermogenic adipocyte differentiation [11] to be specifically expressed in mature BAs ( S4B Fig ) . Finally , we profiled microRNA gene expression along the same process , leading to the identification of known general adipogenic microRNAs ( e . g . miR-378 ) , brown lineage-specific microRNAs ( miR-193 ) , as well as several microRNAs not implicated in adipogenesis so far ( S5 Fig and S2 Table ) . To identify lineage-specific genes , we compared the 431 genes that were stage-specifically expressed in mature BAs ( d7 , Fig 2A ) to those 420 genes that were specifically expressed in mature WAs ( d7 , Fig 2B ) . Among them , we found that 121 genes were robustly expressed specifically in mature BAs but not WAs , and 132 genes were only expressed in mature WAs ( Fig 2C ) . To further compile a list of putative lineage-specific markers for BA and WA , we selected the genes showing a similar lineage-specific expression pattern both in mouse BAT/WAT tissues and primary brown/white adipocytes , according to previously published data [27] ( See S4A Fig for examples ) . The list for BA-specific genes contained a number of classic BA markers , such as Ucp1 , Elovl3 , Ppara and Cidea . In addition , Slc36a2 ( also known as Pat2 ) , recently described as a brown/beige-specific surface marker [28]; Cpt1b , the rate-controlling enzyme for long-chain fatty acid β-oxidation and several other mitochondrial protein genes ( Chcd10 , Sirt3 , Mtfp1 , Aspg and Adssl1 ) were also identified in this list . This observation is consistent with increased number and activity of mitochondria in BAs . Of note , we also found the gene encoding the kinase PIM1 that was specifically expressed in BAs , primary BAs , and BAT . In addition , we also noticed increased Pim1 expression upon cold exposure and chemical activation of the β-adrenergic pathway ( Fig 2D ) . PIM1 belongs to a group of constitutively active serine/threonine kinases and has been implicated in a number of biological functions such as apoptosis and cell cycle regulation . Recent reports suggested that PIM1 might play a role in cellular metabolism by modulating the phosphorylation status of AKT [29] and AMPK [30] . Based on this evidence , we selected PIM1 for further functional analysis as a potential regulator of BA differentiation ( detailed below ) . For WAs , only few markers were established before , of which we rediscovered Nrip1 ( also known as Rip140 ) , a co-repressor that plays an important role in repressing a number of brown selective genes [31] . Another WA-specific gene , Trem2 , was recently shown to enhance adipogenesis , promote glucose and insulin resistance , and diminish energy expenditure . Several other genes identified , such as the nuclear receptor Ear2 ( Nr2f6 ) , the ubiquitin gene Ubd ( Fat10 ) , the free fatty acid receptor Ffar2 , and the insulin-like growth factor Igf1 , were shown to be involved in metabolism without a clear role in white adipogenic differentiation . Finally , we also provide a list of lineage-specific lncRNAs ( S2 Table , see S4B Fig for examples ) . Together , this transcriptomic dataset provides a valuable resource for the identification and further characterization of novel regulators for brown and white adipogenesis . To ask how the lineage-specific and the commonly expressed genes in BA and WA are regulated at the chromatin level , we examined the histone modification dynamics at the gene promoters throughout both BA and WA differentiation . As expected , chromatin marks for active promoters such as H3K4me3 and H3K27ac correlate well with gene activity at the promoter regions in both lineages . Given that a recent study [32] suggested that the removal of H3K27me3 is required for brown gene expression , it was not surprising to observe a decrease of this mark at the promoters of brown specific genes in BA ( Fig 2E ) . Surprisingly , in WA , where these BA selective genes were not expressed , we also found a significant decrease in H3K27me3 at their promoters ( Fig 2E and S6A Fig ) , suggesting that the removal of H3K27me3 is not sufficient to induce the expression of these brown specific genes . Intriguingly , in contrast to WA , we found significantly higher levels of H3K4me1 at the promoters of BA specific genes throughout BA differentiation ( Fig 2E and S6B Fig ) . This observation suggested that the pre-deposition of H3K4me1 at the early stages of brown adipogenesis was required for efficient expression of these genes at the late stage , while the removal of H3K27me3 was necessary but not sufficient to promote brown gene expression . In parallel , we found that general adipogenic genes are only marked by H3K4me1 but not H3K27me3 during both brown and white adipogenesis , suggesting their activation does not involve H3K27 demethylation . Enhancers are cis-regulatory elements that can activate gene expression over distance . It has been shown that enhancers are highly dynamic and play an important role in cell fate transitions [21] . To examine the dynamic regulation of enhancers during BA and WA differentiation , we identified stage-specific enhancers based on the enrichment of H3K27ac , a histone mark for active enhancers and promoters [33] , and the lack of H3K4me3 , a histone mark present at active promoters . We employed a similar entropy-based method as for the identification of stage-specific genes and found 24 , 002 and 13 , 429 genomic loci acting as putative stage-specific enhancers throughout BA and WA differentiation ( Fig 3A and 3B ) . Again the higher number of dynamic enhancers in BA is in agreement with the additional commitment step in the differentiation of MSCs into the brown lineage , as compared to white adipogenesis starting from committed 3T3-L1 preadipocytes . Moreover , we observed the emergence of a distinct group of stage-specific enhancers after BMP7 treatment ( compare d-3 to d0 , Fig 3A ) , which suggested an epigenomic reprogramming during the process of brown lineage commitment . Consistent with a previous report [23] , we also observed an epigenomic transition as evidenced by the formation of a new group of stage-specific enhancers within 6 hours after adipogenic induction ( compare d0 to 6h , Fig 3A ) , while from d2 to d7 , there are less changes as compared to the earlier stages ( Fig 3A ) . During white adipogenesis , the stage-specific enhancers at the early ( d-2 , d0 ) and late stages ( d2 , d7 ) show a certain overlap , while between d0 and d2 , there is a relatively more drastic transition in enhancer formation . This pattern can be explained as at d-2 and d0 , 3T3-L1 cells are under the growth arrest state [25]; after adipogenic induction , these cells go through clonal expansion between d0 and d2 [25]; beyond d2 , the epigenomic reprogramming has been completed and the cells start to accumulate fat and subsequently enter the end differentiation stage at d7 . To validate our analysis , we surveyed the levels of H3K4me1 , another enhancer-associated epigenetic mark , at those loci and found a high concurrence between H3K27ac and H3K4me1 ( Fig 3A and 3B and S1 Table ) . Analysis of the genes present in the proximity of stage-specific enhancers showed that they fall into similar GO categories as the stage-specific genes analyzed earlier . This observation suggests that the stage-specific gene expression was likely regulated by the stage-specific enhancers , further confirming the role of enhancers in cell fate transitions [21] . Enhancers often serve as hubs for TFs . To identify potential TFs that bind to these stage-specific enhancers , we carried out motif analysis of enhancer associated sequences . De novo motif search led to the identification of several enriched binding motifs at each stage of BA and WA differentiation ( S7 Fig ) . While some of the motifs are closely related to known TF binding motifs , ( e . g . PPARγ in mature BA and WA ) most of the motifs could not be assigned to known TFs due to our limited knowledge of the DNA binding motif for most TFs . Therefore , we examined these enhancers for the enrichment of known TF binding motifs derived from previous genome wide TF binding studies [34] . TFs with enriched motifs and robust expression at the corresponding stages are shown in Fig 3C and 3D . We found that motifs for well-known adipogenic regulators such as PPARγ , RXR , C/EBPα and FOXO1 were highly enriched in mature brown as well as white adipocytes . At the early stages , motifs for early adipogenic regulators including PBX1 , KLF4 and STATs were enriched in either white or brown lineages ( Fig 3C and 3D ) . Interestingly , the binding motif for SIX1 , a homeobox transcription factor not previously implicated in the development of BAs was significantly enriched in mature BAs , but not WAs , suggesting a role for this factor in BA differentiation . To validate our finding in an in vivo setting , we also analyzed active enhancers in BAT and WAT tissue ( using previously released datasets from ENCODE [35] ) for enrichment of the SIX1 motif . Indeed , the SIX1 motif was found to be enriched at a much higher level in BAT than in WAT ( S3 Table ) . It has been shown that key cell identity genes are often associated with super-enhancers ( SEs ) [36] , a cluster of enhancers that are enriched for binding of TFs , mediator , and chromatin marks such as H3K27ac [14] . To search for novel regulators of BA differentiation , we sought to map the SEs and define SE-associated genes in both brown and white lineages and identify common as well as lineage specific SE genes . We employed the H3K27ac ChIP-seq data to define SEs because this allowed us to monitor SEs throughout the whole process of BA as well as WA differentiation and determine the SEs present specifically at the late ( d2 and d7 ) but not the earlier stages ( d-3 to 6h ) . To identify genes which are potentially regulated by these SEs , we filtered for those ( 1 ) within 100 kb of the SE and ( 2 ) whose gene expression patterns correlated with SE occurrence throughout differentiation . Through this approach , we identified 419 SE-associated genes for mature BAs and 417 SE genes for mature WAs ( S4 Table ) , of which 324 were BA selective and 322 were WA selective ( Fig 3E ) , respectively . As expected , well-known general adipogenic marker genes such as Cebpa , Fabp4 and Scd were associated with SEs in both lineages . And SE genes at late stages of BA differentiation included most key regulators of brown adipogenesis ( Ucp1 , Cidea , Fgf21 and Ppara ) ( Fig 3E and S8A Fig ) . Moreover , these genes tended to get transcriptionally activated ( S8B Fig ) . Notably , Fabp3 , Pdk4 , Cpt1b and Cpt2 , which were recently identified as putative SE associated genes in a human cell culture model of browning [10] , were associated with SEs only in brown cells . Intriguingly , the TF RREB1 whose gene locus has been linked to metabolic traits like T2D susceptibility , fasting glucose levels , and body fat distribution [37–39] through genome-wide association studies ( GWAS ) , was up-regulated during brown adipogenesis and also associated with a SE defined by H3K27ac enrichment ( Fig 3F ) . This SE encompasses the entire promoter region of Rreb1 . Using PPARγ binding as alternative method to define SEs in mature BAs , we found that Rreb1 was associated with one of the top SEs in BAs due to robust PPARγ binding upstream of its transcriptional start site . In addition to Rreb1 , other SE genes determined by PPARγ binding signals include a whole panel of key brown cell markers such as Ucp1 , Pgc1a , Cidea , Fgf21 and Ppara , and interestingly , Pim1 ( S8C Fig ) . Based on the above observations , we selected Rreb1 for further functional analysis in BA development and function ( detailed below ) . BMP7 strongly promotes brown lineage commitment and differentiation in vitro and in vivo [24] . However , the detailed molecular mechanism underlying BMP7 function and its downstream targets during BA lineage commitment have not been thoroughly characterized . Therefore , we profiled the epigenomic and transcriptomic landscape in C3H10T1/2 cells treated with or without BMP7 . To determine the molecular targets of BMP7 , we compared the transcriptomic profiles between BMP7 treated and untreated C3H10T1/2 cells . On top of that , we also included the corresponding dataset from 3T3-L1 cells for comparison . To identify potential BMP7 targets , we focused on a specific group of 89 genes which showed a robust but transient induction after 3 days of BMP7 treatment ( Fig 4A ) . Amongst those genes are modulators of WNT ( Fzd9 , Frzb ) and TGFβ ( Bambi , Scube3 ) signaling pathways , which were known to be involved in regulating adipogenesis ( see S5 Table ) . In addition , the single most interesting group of genes consisted of two members of the SOX family of transcription factors , Sox8 and Sox13 . Using slightly relaxed cutoff criteria , we noticed that five out of the total 20 Sox genes behave as putative BMP7 targets: Sox5 , 6 , 8 , 9 and 13 ( Fig 4B ) . This observation suggested that at least part of the BMP7 response is mediated by the action of SOX proteins . Sox genes have been implicated in the regulation of embryonic development and in the determination of cell fate [40] . Sox9 expression in rat MSCs increases Cebpb expression and favors adipogenesis [41] and Sox5 , 6 , and 9 play important roles in chondrogenesis . Consistent with this , we observed a transient boost of chondrogenic gene expression after BMP7 treatment and enrichment of related GO categories ( Fig 4C ) . BMP7 activates two major signaling pathways , the SMAD , and the p38 MAPK pathways . Previous studies suggested that p38 signaling is more important for the formation of thermogenic cells and the activation of β-adrenergic pathway [24 , 42] . To determine if Sox genes are downstream targets of either the SMAD or the p38 pathway , we pre-incubated C3H10T1/2 cells with p38 inhibitors prior to treatment with BMP7 , and surveyed their expression . As shown in Fig 4D , Sox gene activation was strictly dependent on p38 signaling as the treatment of p38 inhibitors PD169316 ( PD ) or SB202190 ( SB ) completely abolished the activation of all five Sox genes by BMP7 . We also examined 27 additional genes from the list of BMP7 targets and found their expression was also dependent on p38 signaling ( Fig 4E ) , which seemed to be the major transmitter of the BMP7 signal . Notably , at least two of those genes ( Col11a2 and Col9a1 ) are well-known targets of Sox9 . From the list of five Sox genes , we selected Sox13 for further functional studies ( detailed below ) . In another effort to identify relevant targets from our list of 89 candidates , we defined SEs according to the H3K27ac ChIP-seq data and generated a list containing genes associated with SEs at d0 upon BMP7 treatment . When we intersected both lists we found that 14 BMP7 induced genes were indeed associated with SEs ( S9A Fig ) and those genes might constitute another set of important targets of the cellular response to BMP7 activation . Amongst them , we found the fibroblast growth factor receptor Fgfr3 , which is one of the receptors for FGF21 that promotes both BAT activation and subcutaneous WAT ( scWAT ) browning [43] . In addition to its robust induction by BMP7 , we also observed significantly elevated levels of H3K4me1 , H3K9ac and H3K27ac at the upstream enhancer region and increased H3K4me3 at the promoter of Fgfr3 gene ( S9B Fig ) , underscoring its epigenetic regulation upon BMP7 treatment . Using different approaches of bioinformatics analysis , we identified a number of putative regulators for brown adipogenesis . From these candidates , we selected and validated the following four factors: ( i ) the kinase PIM1 , found to be lineage-specifically expressed in mature BAs but not WAs ( Fig 2C and 2D ) , and three TFs , ( ii ) SIX1 , of which the binding motif was enriched in late stage BA enhancers ( Fig 3C ) , ( iii ) RREB1 , associated with a SE in BAs ( Fig 3F and S8C Fig ) , and finally ( iv ) SOX13 , which is transiently induced by BMP7 during brown lineage commitment ( Fig 4B ) . We first performed gain-of-function analysis of those factors during brown adipogenesis , by lenti-virally over-expressing the corresponding genes in the MSC line C3H10T1/2 before adipogenic induction without BMP7 treatment . As shown in Fig 5A , over-expression of each candidate , or EBF2 , a known regulator of brown adipogenesis [16] , significantly increased the differentiation efficiency of the C3H10T1/2 cells as monitored by Oil-Red-O ( ORO ) staining . Importantly , we also detected higher levels of brown / mitochondrial marker gene expression in the cells over-expressing the four candidate genes . These genes include the brown cell key regulators Prdm16 and Ppara , genes involved in brown cell function ( Cidea and Elovl3 ) , and genes essential for mitochondrial activity ( Cox7a1 and Cox8b ) ( Fig 5B ) . Moreover , Ucp1 , the key thermogenic gene in BAs , was also up-regulated in cells expressing the four candidate genes with or without forskolin treatment ( Fig 5C ) . And these mRNA expression changes were reflected at the protein levels as both CIDEA and PPARα proteins were up-regulated by the over-expression of the four candidates ( Fig 5D ) . In parallel , we also detected increased expression of general adipogenic genes such as Pparg2 , Fabp4 and CD36 upon over-expression of the candidate genes , or Ebf2 ( Fig 5E ) . This observation was consistent with increased lipid accumulation in the corresponding cells as determined by ORO staining ( Fig 5A ) . Increased mitochondria activity leads to up-regulated oxygen consumption rate ( OCR ) , and this is a key feature of thermogenic brown cells . To examine the effects of Pim1 , Six1 , Sox13 , and Rreb1 over-expression on mitochondria activity , we measured the OCR in the corresponding cells 7 days after adipogenic induction . We found that both OCRs ( Fig 5F ) and other cellular metabolic parameters including basal respiration , proton leak , ATP production and maximal respiration ( Fig 5G ) were significantly increased upon candidate over-expression . In addition , we also noticed a shift towards uncoupled respiration ( Fig 5H ) , suggesting enhanced thermogenesis . The chemical activation of the β-adrenergic pathway by drugs such as norepinephrine can significantly stimulate BAT activity . To examine the effects of candidate over-expression on the response to β-adrenergic activation , we measured the OCR in corresponding cells after norepinephrine treatment . The results showed that cells over-expressing Pim1 , Six1 , Sox13 , Rreb1 , or Ebf2 ( Fig 5J ) were more susceptible to β-adrenergic activation than control cells ( Fig 5I ) , indicating enhanced thermogenic capability . Taken together , these results suggested that all four candidates either facilitate the commitment or the differentiation process from MSCs to functional BAs . We showed that Pim1 , Six1 , Sox13 and Rreb1 were sufficient to promote brown adipogenesis . To test whether they were also necessary for BA differentiation , we performed loss-of-function analysis using Stromal Vascular Fraction ( SVF ) cells isolated from BAT transfected with LNA longRNA GapmeR oligonucleotides to knock down the genes of interest before adipogenic induction . As shown in Fig 6A , cells transfected with a scramble oligo ( Scr ) readily differentiated into mature BAs , whereas knock-down of either the candidate genes using two independent GapmeRs or Pparg led to severely reduced capabilities to differentiate as demonstrated by ORO staining . In addition , prominent brown / mitochondrial regulators and markers such as Prdm16 , Pgc1a , Ppara , Cidea , Elovl3 , Cox7a1 and Cox8b ( Fig 6B ) , as well as Ucp1 ( Fig 6C ) were down-regulated in cells with Pim1 , Six1 , Sox13 , and Rreb1 knock-down . Consistent with the ORO staining results , adipogenic markers including Pparg2 , Fabp4 and CD36 were also reduced ( Fig 6D ) by the knock-down of the candidate genes ( Fig 6E ) . We also validated the function of the four candidates in SVF cells isolated from posterior scWAT . Those cells have a certain capacity to “brown” [44] and over-expression or knock-down of the four candidates resulted in similar outcomes as observed in the MSC and BAT SVF cell systems ( S10 and S11 Figs ) . To gain further mechanistic insight into the mode of action for one of the identified factors , SIX1 , we mapped its genomic localization via ChIP-seq in mature BAs . We found in total 7366 binding peaks for SIX1 with most of them located at intergenic regions , introns and promoters ( Fig 7A ) , which is typical for TFs [45] . GO analysis of the SIX1 binding genes revealed that “regulation of generation of precursor metabolites and energy” , “negative regulation of TGFβ receptor signaling pathway” and “brown fat cell differentiation” were amongst the most significantly enriched categories ( Fig 7B ) . We detected SIX1 binding at the cis-regulatory regions ( marked by H3K27ac ) of brown markers such as Cidea and Ucp1 ( Fig 7C ) . Moreover , we observed partial overlap of SIX1 binding to PPARγ binding at these regions . In a more quantitative analysis we measured the strength and proximity of SIX1 binding at brown-specific , white-specific and commonly expressed ( i . e . white and brown ) genes . We found that SIX1 bound preferentially around brown-specific and commonly expressed genes as compared to white-specific genes , suggesting a role for this factor in regulating brown selective as well as general adipogenic gene expression ( Fig 7D ) . To decipher the molecular mechanism underlying SIX1 function , we performed a motif analysis of SIX1 bound regions . As expected , the most enriched binding motif was for SIX1 itself , which was followed by motifs for C/EBP , EBF , and NF1 TFs ( Fig 7E ) . PPARγ and RXR binding motifs were only mildly enriched . The enrichment of C/EBP and EBF binding motifs at SIX1 binding sites suggested physical interactions between these TFs . Indeed , we verified the direct interactions between SIX1 and C/EBPα , C/EBPβ , as well as EBF2 using co-IP assays ( Fig 7F ) . Finally , using luciferase activity assay , we found that an upstream enhancer element of the Cidea gene harboring a SIX1 motif ( S12 Fig ) promotes expression in a SIX1-dependant manner ( Fig 7G ) . Together , our findings corroborate a model in which SIX1 can be recruited to brown-specific or general adipogenic genes through either direct DNA binding ( via the SIX1-binding motif ) or recruitment by EBF2 and C/EBP proteins ( at regions with no SIX1-binding motif ) .
Promoting energy expenditure through thermogenesis is of significant interest as potential therapy for obesity and related diseases . It requires the recruitment of thermogenic fat cells such as brown and beige/brite adipocytes . Existing evidence suggests that the majority of these thermogenic cells are recruited de novo in response to environment cues [43 , 46–48] . Therefore , to promote thermogenic adipocyte recruitment , it is essential to have a fundamental understanding of the gene regulation network that governs brown adipogenesis , especially at the lineage commitment step . In this study , we provide comprehensive profiles of the transcriptome and epigenome at five key developmental stages throughout the differentiation of murine multi-potent MSCs into mature BAs . Through in-depth bioinformatics analyses , we identified and functionally validated PIM1 , SIX1 , RREB1 , and SOX13 as novel regulators promoting brown cell differentiation and function . Differential gene expression analysis is a classic approach for the identification of regulators of cell type specification . A number of adipogenic and brown fat cell regulators including PPARγ , C/EBPα and PRDM16 were identified through this approach . In our study , we also used this analysis to identify brown selective genes but added additional criteria for the selection of candidates: these genes must be dynamically regulated during adipogenesis and stage-specifically expressed only in mature adipocytes ( Fig 2 ) . As the result , our list of 121 brown selective genes contains brown markers such as Cidea ( #1 ) , Elovl3 ( #3 ) , Ucp1 ( #24 ) and Ppara ( #61 ) , as well as a number of mitochondrial genes including Cpt1b ( #4 ) . From this list , we specifically looked for factors that could potentially be involved in gene regulation or signal transduction , and we selected the kinase PIM1 ( #53 ) for further analysis . Moreover , the Pim1 gene was later found to be associated with a SE in brown cells ( S8C Fig ) . In our study , over-expression of Pim1 in both C3H10T1/2 cells and the scWAT SVF cells up-regulated a number of key brown cell marker genes as well as general adipogenic genes ( Fig 5B–5E and S10B–S10D Fig ) . In addition , over-expression of this kinase also promoted the mitochondrial respiration in general and specifically uncoupled respiration , a feature of thermogenic fat cells ( Fig 5F–5H and S10F and S10G Fig ) . In contrast , knock-down of Pim1 by GapmeRs reduced the expression of brown marker genes and adipogenesis efficiency in both primary brown cells ( Fig 6 ) and subcutaneous white cells ( S11 Fig ) . Therefore our analysis clearly implicates Pim1 in brown adipogenic differentiation , although future experiments will have to address if its role is solely restricted to the brown lineage . With our experimental model we cannot rule out that enhanced brown differentiation is partially caused by an increase in adipogenic differentiation in general . However , the lineage specific expression of Pim1 in BAT vs WAT , together with its increased expression in BAT upon cold exposure ( Fig 2D ) , point towards a more specific role in BAT for this kinase . It will be interesting to investigate the functional significance and molecular mechanism of PIM1 in different tissues and under different metabolically challenging conditions such as diet induced obesity or cold exposure in vivo , especially the direct targets of this kinase . Enhancer binding motif analysis is another powerful tool to identify novel TFs involved in specific cell differentiation processes [15 , 16] . In our study , we first defined stage-specific enhancers during both white and brown adipogenesis , then surveyed the enrichment of TF binding motifs at early and late stages of differentiation . While enrichment for several well-known adipogenic factors at late stages of both brown and white adipogenesis was expected , our finding that the motif for the TF SIX1 was enriched during late brown adipogenesis was surprising ( Fig 3C ) , since SIX1 has not been implicated in brown cell differentiation so far . Six1 belongs to the Sine Oculis Homeobox family of genes and has been reported to play a crucial role in muscle cell lineage decision as well as muscle development . Through gain- and loss-of-function assays , we confirmed that Six1 was required for the expression of brown selective and adipogenic marker genes in both the C3H10T1/2 cells and the SVF cells from scWAT and BAT . Moreover , over-expression of Six1 enhanced the mitochondrial uncoupled respiration in differentiated C3H10T1/2 cells ( Fig 5H ) . In an attempt to decipher SIX1’s mode of action , we performed genome-wide binding profiling of SIX1 in mature BAs . Strikingly , we found that SIX1 bound to brown as well as general adipogenic genes , some of whose expression were affected by the modulation of the Six1 gene , suggesting that these genes are direct targets of SIX1 . Through binding motif analysis of SIX1 occupied regions and subsequent co-IP assay , we confirmed that SIX1 directly interacts and may cooperate with C/EBPα , C/EBPβ and EBF2 in regulating the transcription network during differentiation . Therefore the role of SIX1 seems not to be restricted to the regulation of BA differentiation , but it may act as a more general activator of the adipogenic transcriptional program . Again , in vivo studies to investigate the physiological role of Six1 in different adipogenic tissues will be instrumental to fully understand its biological function . Super-enhancers are big clusters of enhancers and are often associated with genes specifying cell identity [14] . It can be defined by either mediator or TF binding or the enrichment of chromatin marks such as H3K27ac . Through analysis of super-enhancer associated genes , KLF11 was identified as an important factor promoting the browning of human mature white fat cells [10] . In our study , we profiled the stage-specific super-enhancers in a distinct process: i . e . the differentiation of brown adipocytes from multi-potent progenitor cells . That way we identified 419 genes associated with mature BA specific super-enhancers , including most brown marker genes . From this gene list , we selected the TF RREB1 for further functional analysis on the basis of its link to metabolic traits and association with one of the highest ranking SEs ( #8 ) in BAs as defined by PPARγ binding . In our functional studies , over-expression of Rreb1 led to increased expression of brown marker genes and enhanced mitochondrial respiration in both C3H10T1/2 cells and SVF cells from scWAT . On the other hand , knock-down of Rreb1 resulted in reduced brown marker expression and impaired adipogenesis of SVF cells from BAT and scWAT . Moreover , RREB1 was very recently identified as a positive regulator of brown adipogenesis via a distinct bioinformatics approach [32] during the preparation of this manuscript . Further physiological studies on the role of Rreb1 in insulin sensitivity and energy homeostasis using corresponding gain- and loss-of-function animal models will be necessary to fully characterize the function of this positive regulator of brown adipogenesis . BMP7 was recently established as a key factor that specifies the brown lineage from MSCs [24] . Mechanistically , BMP7 acts through either the SMAD or the p38 MAPK signaling pathway to induce its downstream target genes governing the adipogenic or thermogenic program . In order to gain an in-depth understanding of the brown lineage commitment from multi-potent progenitor cells ( as exemplified by BMP7 signaling ) , we systematically compared the gene expression profiles of C3H10T1/2 cells with or without BMP7 treatment . We found that 89 genes were transiently induced by BMP7 treatment , amongst which we identified a panel of Sox genes . We found that those genes were all p38/MAPK dependent , suggesting their involvement in promoting the thermogenic rather than the general adipogenic program . Subsequent functional analysis confirmed that one candidate gene , Sox13 , promotes adipogenic differentiation , brown marker gene expression , and mitochondrial respiration . Our current results suggest that BMP7 triggers an early p38 dependent response , including the activation of Sox gene expression , important for the lineage commitment of brown cells from multi-potent progenitor cells . Clearly , future work is needed to dissect the exact contribution of the different Sox genes during this process and to define their downstream target genes . Transcriptomic profiles and epigenomic landscapes are important resources for understanding the gene regulation network in a certain cell type or in a specific cell differentiation process . Analysis of those datasets has led to the identification of numerous novel regulators in various cellular processes . Seminal works in the field have provided valuable resources for further investigation of the molecular control of fat cell differentiation [10 , 15 , 23 , 49–53] . Here we add a comprehensive study of the epigenomic and transcriptomic transitions at five key developmental stages throughout the process of murine brown adipogenesis . Our dataset comprises a high temporal resolution of the differentiation process as well as a broad coverage of chromatin marks . Through comparative analyses of white and brown datasets using various bioinformatics tools , we identified many potential candidates and validated four factors that promote thermogenic adipocyte differentiation in various cellular models including C3H10T1/2 cells , SVF cells from subcutaneous WAT and interscapular BAT . Moreover , through analyzing the chromatin dynamics at the promoters of lineage-specific and commonly expressed adipogenic genes in BA and WA , we found that , in addition to the mechanism proposed by a recent study [32] , which suggested that the removal of H3K27me3 is required for brown gene expression , the pre-deposition of H3K4me1 at these genes during early stages of brown adipogenesis is essential for poising them for expression at a later stage . For general adipogenic genes , we found they are only marked by H3K4me1 but not H3K27me3 during both brown and white adipogenesis , suggesting their activation does not involve H3K27 demethylation . Based on these observations , we propose that the pre-deposition of H3K4me1 at brown specific genes is a critical step in the chromatin remodeling during the process of brown adipocyte lineage commitment , while the full activation of these genes is only possible once the “stop sign” ( H3K27me3 ) is removed . Besides proposing this conceptual model for the epigenetic regulation of brown lineage specific genes , we anticipate additional factors ( including lncRNAs and miRNAs ) , promoting or inhibiting brown cell differentiation to be identified through analyzing these datasets and further insights to be gained from these resources .
3T3-L1 preadipocytes and C3H10T1/2 mesenchymal stem cells were purchased from ATCC . 3T3-L1 cells were maintained in DMEM ( Gibco , 11995–065 ) supplemented with 10% bovine calf serum ( BCS; HyClone , SH30072 . 03 ) . For 3T3-L1 differentiation , BCS was replaced by 10% fetal bovine serum ( FBS; HyClone , SH30070 . 03 ) and cells were seeded on gelatinized dishes to reach 70% confluency on the next day ( d-4 ) . Two days later cells reached full confluency ( d-2 ) and after another two days ( d0 ) differentiation was induced by adding 1 μM dexamethasone ( Sigma , D4902 ) , 0 . 5 mM 3-isobutyl-1-methylxanthine ( Sigma , I7018 ) , and 10 μg/ml insulin ( Santa Cruz , sc-360248 ) for two days . Subsequently cells were maintained in DMEM with 10% FBS and 10 μg/ml insulin; medium was changed every other day . To differentiate C3H10T1/2 cells into brown adipocytes , cells were maintained in DMEM ( high glucose ) supplemented with 10% Fetal Clone III serum ( Hyclone , SH 30109 . 03 ) , split onto gelatinized dishes at 70% confluency the next day ( d-3 ) and treated with 8 . 3 nM human recombinant BMP7 ( R&D Systems , 354-BP ) for three days . At day 0 , differentiation was induced by adding 5 μM dexamethasone , 0 . 5 mM 3-isobutyl-1-methylxanthine , and 0 . 12 μg/ml insulin , 1 μM rosiglitazone ( Cayman , 71740 ) and 1 nM 3 , 3’ , 5-Triiodo-L-thyronine ( Sigma , T5516 ) to the medium . From day two on the cells were maintained in DMEM , 10% FetalClone III with 0 . 12 μg/ml insulin , 1 μM rosiglitazone and 1 nM 3 , 3’ , 5-Triiodo-L-thyronine . The medium was replaced every two days . To test the effect of p38 inhibition , 10 μM PD169316 ( GenEthics ) or 10 μM SB202190 ( Sigma ) was added to the medium six hours prior to BMP7 treatment for three days . SVF cells were isolated from 8 weeks old male C57BL/6J mice . Brown adipose tissue from the interscapular region or posterior subcutaneous white adipose tissue was collected , rinsed in 1x HBSS ( Gibco , 14175 ) supplemented with 50 μg/ml D-Glucose ( Sigma , G8644 ) , cut into small pieces , digested with 3 ml collagenase solution ( Collagenase 1 mg/ml , Sigma , C9891; BSA 20 mg/ml , Sigma-Aldrich , A7906; D-Glucose 50 μg/ml , Sigma , G8644; in 1xHBSS ) per 1 g of tissue for 1–1 . 5 hours by nutating at 37°C . The digestion was stopped by adding SVF growth medium ( DMEM/F12 , Gibco , 11330–032; 20% FBS , Hyclone; PenStrep , Gibco , 15140–122 ) . Cells were pipetted up and down carefully , centrifuged for 5 min at RT at 400g . After careful resuspension of cells in fresh growth medium , they were passed through a 100 μM filter mesh ( BD Falcon , 352360 ) , centrifuged again , and resuspended in 1xHBSS supplemented with 50 μg/ml D-Glucose . The washing step was repeated twice . After the last wash , cells were resuspended in 1x Red Blood Lysis buffer ( Biolegend , 420301 ) , incubated for 5 min at RT before another centrifugation . Finally , cells were resuspended in growth medium and plated onto cell culture plates . Cells were expanded in growth medium and differentiation started not later than passage four . For adipogenic differentiation , cells were seeded on gelatinized 12-well plates ( 40 , 000 cells/well ) in differentiation medium ( DMEM/F12; 10% FBS ) . Two days later the medium was replaced by induction medium ( DMEM/F12; 10% FBS; 5 μM dexamethasone , 0 . 5 mM 3-isobutyl-1-methylxanthine , 0 . 5 μg/ml insulin , 1 μM rosiglitazone , and 1 nM 3 , 3’ , 5-Triiodo-L-thyronine ) . After three days the medium was changed to differentiation medium supplemented with 0 . 5 μg/ml insulin and 1 nM 3 , 3’ , 5-Triiodo-L-thyronine and replaced every other day . Plasmids containing full length cDNA were purchased from OriGene ( hSix1 ( RC203465 ) , hSox13 ( RC210697 ) , hPim1 ( RC205853 ) , mEbf2 ( MR224591 ) ) or Addgene ( hRreb1 ( 41145 ) ) , and sub-cloned into a lenti-viral vector providing an EF1a promoter for expression and a C-terminal tGFP-tag ( OriGene PS100072 ) . In order to not exceed the packaging capacity of the lenti-virus , the C-terminal portion of hRreb1 ( the last 333 amino acids ) was removed , leaving behind a truncated version of Rreb1 , hRreb1ΔC ( 1–1408 amino acids ) . Lenti-viral particles were produced in HEK293NT cells using the Lenti-vpack lenti-viral packaging kit from OriGene ( TR30022 ) following the manufacturer’s protocol . C3H10T1/2 cells or SVF cells were transduced with lenti-viral supernatant diluted 1:1 in fresh growth medium and 8 μg/ml Polybrene overnight . The following LNA-longRNA GapmeRs ( 300600 ) from Exiqon were used to knock down the indicated genes: mSix1 #1 ( CAAACTGGAGGTGAGT ) , mSix1 #2 ( CAGAGGAGAGAGTTGA ) ; mSox13 #1 ( GCAAAGGCTGGTGGCT ) , mSox13 #2 ( GAGGAGGAGGTTTAGC ) ; mPim1 #1 ( GGAGTTGATCTTGGAC ) , mPim1 #2 ( GGTGATAAAGTCGA ) ; mRreb1 #1 ( GTTAGATTTGGTAGA ) , mRreb1 #2 ( CGTTGATGAGAGGTG ) ; Pparg ( AGAAATCAACTGTGGT ) ; scr ( /56-FAM/AACACGTCTATACGC ) . Prior to transfection , the SVF cells were seeded on gelatinized 12 well plates ( 40 , 000 cells/well ) . On the next day , cells were transfected with 60 μM LNA-longRNA GapmerRs using 4 . 5 μl Lipofectamine 2000 ( Invitrogen , 11668019 ) . To determine cellular oxygen consumption rates , we used the Seahorse XFe24 Extracellular Flux Analyzer . Cells were seeded on gelatinized XFe24 cell culture microplates ( 100777–004 ) at 4 , 000 cells/well and differentiated one day post confluency following the procedures described above . The XF Cell Mito Stress Test Kit ( 103015–100 ) was used to determine basal respiration , ATP production , proton leak , maximal respiration , and spare respiratory capacity . Concentrations of the added chemicals were: 40 μM oligomycin , 0 . 15 μM FCCP , 1 μM rotenone / 1 μM antimycin A . To activate the β-adrenergic pathway before measurement , cells were treated with 10 μM norepinephrine ( Sigma Aldrich , A0937 ) . Cells were harvested with TRIzol reagent ( Ambion , 15596018 ) and total RNA was extracted following the manufacturer’s protocol . RNA was treated with Amplification Grade DNase I ( Invitrogen , 18068–015 ) before the generation of either RNA-seq libraries or cDNA to assess expression of individual genes . RNA-seq libraries were generated by BGI , China , following standard procedures . For cDNA generation , 500 ng of total RNA were reverse transcribed using random 9-mers and M-MLV Reverse Transcriptase ( Invitrogen , 28025–021 ) . Target gene expression was determined via quantitative Real-Time PCR analysis using Power SYBR Green PCR Master Mix from Applied Biosystems ( 4367659 ) on a 7900HT Fast Real-Time PCR machine ( Applied Biosystems ) . All gene expression data in this study was normalized to the expression of the riboprotein gene 36B4 unless indicated differently . Primers were listed in S6 Table . Chromatin immunoprecipitation was performed as described earlier [54]; antibodies used in this study were listed in S6 Table . To construct ChIP-seq libraries , we employed a method described previously [55] . In short , 5ng of ChIP DNA were used as starting material . After an end repair step , terminal addition of poly-dCs , and ligation of linkers , the DNA was amplified in a two-step PCR procedure . The final product was loaded onto a 2% agarose gel and fragments between 200bp and 500 bp were cut from the gel , purified and sequenced at BGI China . Cells were lysed using a modified version of the RIPA buffer ( 50 mM Tris-HCl pH8 , 0 . 5% NaDeoxycholate , 150 mM NaCl , 1% NP-40 , 0 . 1% SDS ) supplemented with proteinase inhibitor cocktail mix ( Roche , 11 873 580 001 ) , sonicated ( Bioruptor , 10 cycles , 10 seconds , high energy ) , and centrifuged for 15 minutes at 4°C . The cell lysates ( supernatants ) were quantified using Bradford reagent ( Bio-rad , 500–0205 ) , separated on the Bio-rad Minigel system , and blotted onto nitrocellulose membranes ( Bio-rad , 162–0115 ) , which were blocked using StartingBlock T20 blocking buffer from Thermo Scientific ( 37543 ) . Antibodies used were listed in S6 Table . Signals were detected using the WesternBright ECL detection kit from Advansta ( K-12045-D20 ) . HEK293 cells were transfected using Lipofectamine 2000 ( Invitrogen ) with the following plasmids: pCMV6-hSix1:myc:DDK from Origene ( RC203465 ) ; pcDNA3 . 1-Cebpa:HA , a kind gift from Dr . D . Tenen; pcDNA3 . 1-mCebpb from Addgene ( 12557 ) ; pLenti-EF1ap-mEbf2:tGFP ( Origene ) . Three days after transfection , cells were harvested and lysates were prepared as for Western blots except that the lysis buffer contained 50 mM Tris-HCl pH7 . 4 , 150 mM NaCl , 1 mM EDTA , 0 . 1% NP40 , and proteinase inhibitor mix ( Roche , 11 873 580 001 ) . 500 μg of total proteins were pre-cleared with 10 μl protein G beads ( Sigma , P3296 ) and incubated overnight with 10 μl anti-FLAG M2 affinity gel at 4°C on a rotator ( Sigma , A2220 ) . The gel was washed twice each with lysis buffer , RIPA buffer containing 0 . 1% SDS , RIPA buffer containing 0 . 1% SDS and 1% NP-40 . Precipitated proteins were eluted using SDS loading buffer ( 95°C , 5 min ) and separated on PAGE gels for Western analysis . A 222 bp fragment corresponding to the region -13 , 982 to -13 , 761 from the Cidea transcription start site ( TSS ) was PCR-amplified from mouse genomic DNA and cloned into pGL3-Basic plasmid ( Promega ) to yield pGL3-Cidea-Enh1 . For the luciferase assay , HEK293 cells were transfected with pGL3-Cidea-Enh1 ( or equal amounts of pGL3-Basic ) and pCMV6-hSix1:myc:DDK from OriGene ( RC203465 ) ( or equal amounts of pCMV6:myc:DDK ) using Lipofectamine 2000 ( Invitrogen ) . 48h after transfection cells were lysed and Firefly luciferase activity was measured on a GloMax Multi luciferase reader using the Dual-Glo luciferase assay kit ( Promega ) following the manufacturer’s recommendations . Cells were washed with 1x PBS , fixed with 10% formaldehyde ( Sigma-Aldrich , 252549 ) for 20 minutes at room temperature , washed three times with 1x PBS , and incubated with freshly made Oil-Red-O working solution ( 60% Oil-Red-O stock solution [5 mg/ml Oil-Red-O ( Sigma-Aldrich , O0625 ) in 60% triethylphosphate solution] and 40% dH2O ) for 1h . Finally , cells were rinsed with dH2O . All ChIP-seq datasets were aligned using Bowtie ( version 2 . 0 . 4 ) to build version mm9 of the murine genome . Alignments were performed with the following additional parameters: -t -q -p 8 -N 1 -L 25 . To visualize the ChIP-seq signals for each histone modification and PPARγ , we extended each read to 300 bp and counted the coverage of each read for each base , which was shown as the UCSC genome browser tracks . For the downstream analysis , we normalized the read counts for the ChIP samples by computing the numbers of Reads Per Kilobase of bin per Million reads sequenced ( RPKM ) . To minimize the batch and cell type variation , the RPKM values were further normalized through Z-score transformation . MACS [56] was used to identify histone modification regions and PPARγ binding peaks by default settings . All RNA-seq datasets were aligned using Tophat ( version 2 . 0 . 6 ) to build version mm9 of the murine genome . Alignments were performed with the following additional parameters:-p 2—solexa1 . 3-quals . The mapped reads were further analyzed by Cufflinks ( v 2 . 1 . 1 ) [57] and the expression levels for each transcript were quantified as Fragments Per Kilobase of transcript per Million mapped reads ( FPKM ) based on refFlat database . To identify differentiation stage-specific coding genes , we used a strategy described previously based on the Shannon entropy to compute a stage-specificity index to all expressed coding genes [58–60] using the averages of two RNA-seq replicates . The entropy score for each gene was defined as described as follows [61]: for each gene , we defined its relative expression in a cell type i as Ri = Ei /∑E , where Ei is the FPKM value for gene expression in the cell type i; ∑E is the sum of FPKM values in all cell types; N is the total number of cell types . Then the entropy score for this gene across cell types can be defined as H = -1*∑Ri * logRi ( 1≤i≤N ) , where the value of H ranges between 0 to log2 ( N ) . An entropy score close to zero indicates the expression of this gene is highly stage-specific , while an entropy score close to log2 ( N ) indicates that this gene is expressed ubiquitously . Based on an examination of the entropy distribution genes with entropy less than 2 were selected as the candidate stage specific genes . Among these candidates , we selected genes for each stage based on the following criteria: the gene was highly expressed in this stage ( FPKM>5 ) , and high expression ( FPKM>5 ) could not be observed in more than three additional stages . These genes were then reported in the final stage specific coding gene list and used for subsequent analyses , e . g . GO analysis ( see below ) . To identify stage specific lncRNAs , we used NONCODE v4 mouse annotation database [26] and removed the regions overlapping with refFlat annotation . The expressed putative lncRNAs are those with FPKM>0 . 5 in at least one of the five differentiation time points for each lineage . We used entropy score less than 2 to select candidates of stage specific lncRNA and applied the following criteria: the lncRNA was highly expressed in this stage ( FPKM>0 . 5 ) , and expression ( FPKM>0 . 5 ) could not be observed in more than three additional stages for this lncRNA . These lncRNAs were then reported in the final stage specific lncRNA list . The microRNA array signals were quantile normalized among different stages during adipogenesis and we selected those with expression varying by more than two-fold between different stages during adipogenesis . They were assigned to their specific stages based on their maximum expression . Day 7 adipocyte specific coding genes were identified from the day 7 stage specific gene list obtained above . We selected the brown d7 adipocyte specific genes based on the following criteria: ( 1 ) brown adipocyte d7 expression was at least 3 fold higher than that in white d7 adipocytes; ( 2 ) this genes was not highly expressed in white d7 adipocyte ( FPKM<10 ) . To generate the list of lineage-specific genes in Fig 2C , we introduced two additional criteria: ( 1 ) expression in differentiated SVFs from BAT was at least 2 . 5-fold higher than in differentiated SVFs from WAT ( data from [27] ) , and ( 2 ) expression in BAT tissue was at least 2 . 5-fold higher than in WAT tissue ( harvested from male C57BL6 mice ) . The criteria for white specific genes were vice versa . Day 7 adipocyte specific lncRNAs were identified from the day 7 stage specific lncRNA list obtained above . To be defined as brown specific lncRNAs , their expression ( FPKM ) at day 7 had to be at least 5-fold higher in brown than that in white adipocytes . The criteria for white d7 adipocyte specific lncRNAs were vice versa . To identify enhancers ( without overlapping promoters ) we first called H3K27ac and H3K4me3 peaks for each stage by MACS . The peaks from all stages ( BA: d-3 , d0 , 6h , d2 , d7; WA: d-2 , d0 , d2 , d7 ) were pooled for individual modifications and lineages . Enhancers were identified from the H3K27ac regions not overlapping with H3K4me3 regions and regions 2 . 5 kb up- and down-stream of Refseq transcription start sites . We calculated the RPKM within these enhancer regions and their normalized signals by Z-score normalization . Stage specific enhancers were defined based on the following criteria: the enhancer was highly active in this stage ( normalized RPKM>1 ) , and its high activity ( normalized RPKM>0 ) could not be observed in more than three additional stages . These enhancers were then reported in the final stage specific enhancer list . We first extracted BA d7 and WA d7 stage specific genes as described above . Those genes were first pooled , and then separated into BA-specific , WA-specific and shared common adipogenic genes using the following criteria: a gene was defined as BA-specific if ( 1 ) brown adipocyte d7 expression was at least 3 fold higher than that in white d7 adipocytes; ( 2 ) this gene was not highly expressed in white d7 adipocyte ( FPKM<10 ) . The criteria for WA-specific genes were vice versa . The rest of the genes were identified as shared genes . Then we calculated the z-score normalized RPKM of each histone modification on promoter regions . Then all these normalized RPKM signals were mapped to its corresponding genes . The box plots show the normalized RPKM of different histone modification on different groups of genes . p-values were calculated using t-test . Super-enhancers of brown adipocyte and white adipocyte were identified by the software ROSE [36] based on our brown adipocyte d7 PPARγ ChIP-seq data and published 3T3-L1 d7 PPARγ ChIP-seq data . The closest gene to the super-enhancer was assigned as its potential target . Alternatively , SEs were identified using the H3K27ac datasets . To do this , we calculated Z-score normalized RPKM values on enhancer regions identified above and selected active enhancers with normalized RPKM >0 within 12 . 5 kb to further merge nearby signals into ‘stitched’ enhancers . We kept the ‘stitched’ enhancers larger than 12 . 5 kb as super-enhancer candidates . The RPKM values were calculated on these new super-enhancer candidates and ranked to find the super enhancers with slope > 0 . 5 . To identify SEs specific to late BAs or late WAs , we used the corresponding d7 SEs and removed SEs which showed an overlap with SEs also present at early stages ( i . e . d-3 , d0 , 6h for BA and d-2 , d0 for WA ) . We assigned super enhancers to expressed genes ( FPKM>1 ) within 100 kb of super-enhancer based on the Pearson correlation between expression and enhancer activity during adipogenesis . Genes with correlation coefficient higher than 0 . 75 were selected as the potential super-enhancer targets . If there was no gene with more than 0 . 75 correlation coefficiency within 100 kb of a super-enhancer , we assigned the gene with the largest correlation coefficiency above 0 . 5 as its potential target . microRNAs within 100 kb of super-enhancer were all assigned as potential targets . To find the sequence motif enriched in the identified enhancers , we used the Homer [34] program based on mm9 genome using default settings . For motif analysis , background sequences were randomly selected from the genome , matched for GC% content to facilitate subsequent GC normalization . To find enriched GO categories , we used the DAVID web-tool ( using default settings ) [62] with Gene Ontology database including Molecular Function terms , Biological Function terms and Cell Component terms . To find the GO categories enriched around enhancers , super-enhancers and TF binding peaks , we used the GREAT tool [63] using default settings . Background calculation was based on the whole genome . Six1 peaks were called by MACS as described above . The overlap between the peaks and annotated genome elements was calculated based on the mm9 genome annotation . To calculate Six1 enrichment , WA-specific , BA-specific and common genes were defined as follows: WA-specific genes: FPKM>5 in WA d7 and FPKM<5 in BA d7; BA-specific genes: FPKM>5 in BA d7 and FPKM<5 in WA d7; common genes show FPKM>5 in both WA d7 and BA d7 . When comparing the Six1 signal around those groups of genes , we first defined an unbiased set of regulatory sites and potential Six1 binding site around the TSS using H3K27ac peaks from brown adipocytes ( d7 ) , H3K27ac peaks from white adipocytes ( d7 ) or shared H3K27ac peaks from both brown and white adipocytes . Overlapping Six1 peaks were counted and enrichment was calculated as RPKM by adding the signals of Six1 peaks . Distances between genes and their closest Six1 binding sites were based on their annotated TSSs . All ChIP-seq , RNA-seq and microRNA microarray datasets were deposited in GEO under the accession number of GSE75698 . All animal procedures were performed according to a protocol ( IACUC#130829 ) approved by the Institutional Animal Care and Use Committee of the Agency for Science , Technology and Research ( A*STAR ) of Singapore . | Obesity and its related metabolic diseases are growing problems worldwide . Brown adipose tissue ( BAT ) with its capability of burning off fat to generate heat is now at the center of research interest as target of therapeutic intervention for obesity treatment . In order to get a better understanding of the molecular mechanisms and transcriptional programs underlying brown adipocyte differentiation , we profiled the epigenomic and transcriptomic changes during brown adipogenesis and performed a comparative analysis against white adipogenesis using bioinformatic tools . We identified several novel factors involved in brown adipocyte differentiation and showed that the kinase PIM1 and the transcription factors SIX1 , SOX13 and RREB1 positively regulate differentiation . Finally we also provide a genome-wide map of SIX1 binding in mature brown adipocytes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"and",
"health",
"sciences",
"chemical",
"characterization",
"gene",
"regulation",
"long",
"non-coding",
"rnas",
"cell",
"differentiation",
"developmental",
"biology",
"regulator",
"genes",
"adipocytes",
"connective",
"tissue",
"cells",
"gene",
"types",
"research",
"and",
"analysis",
"methods",
"adipocyte",
"differentiation",
"animal",
"cells",
"gene",
"expression",
"connective",
"tissue",
"biological",
"tissue",
"binding",
"analysis",
"biochemistry",
"rna",
"cell",
"biology",
"anatomy",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"non-coding",
"rna"
] | 2016 | Comparative Transcriptomic and Epigenomic Analyses Reveal New Regulators of Murine Brown Adipogenesis |
The accumulation of mutant mitochondrial DNA ( mtDNA ) molecules in aged cells has been associated with mitochondrial dysfunction , age-related diseases and the ageing process itself . This accumulation has been shown to often occur clonally , where mutant mtDNA grow in number and overpopulate the wild-type mtDNA . However , the cell possesses quality control ( QC ) mechanisms that maintain mitochondrial function , in which dysfunctional mitochondria are isolated and removed by selective fusion and mitochondrial autophagy ( mitophagy ) , respectively . The aim of this study is to elucidate the circumstances related to mitochondrial QC that allow the expansion of mutant mtDNA molecules . For the purpose of the study , we have developed a mathematical model of mitochondrial QC process by extending our previous validated model of mitochondrial turnover and fusion-fission . A global sensitivity analysis of the model suggested that the selectivity of mitophagy and fusion is the most critical QC parameter for clearing de novo mutant mtDNA molecules . We further simulated several scenarios involving perturbations of key QC parameters to gain a better understanding of their dynamic and synergistic interactions . Our model simulations showed that a higher frequency of mitochondrial fusion-fission can provide a faster clearance of mutant mtDNA , but only when mutant–rich mitochondria that are transiently created are efficiently prevented from re-fusing with other mitochondria and selectively removed . Otherwise , faster fusion-fission quickens the accumulation of mutant mtDNA . Finally , we used the insights gained from model simulations and analysis to propose a possible circumstance involving deterioration of mitochondrial QC that permits mutant mtDNA to expand with age .
Mitochondria are multi-functional organelles of eukaryotic cells . While their main function is to generate ATP through oxidative phosphorylation ( OXPHOS ) , mitochondria also participate in fatty acid oxidation , apoptosis , the cell cycle , and cell signaling [1] . Because of the importance of mitochondria , loss of mitochondrial function is detrimental to the organismal well-being . Mitochondrial dysfunction has been associated with a wide range of diseases , such as cancer , diabetes , presbycusis , sarcopenia , and neurodegenerative diseases [2 , 3] . Mitochondria possess their own genome , mitochondrial DNA ( mtDNA ) , encoding proteins that are involved in OXPHOS [4] . A single eukaryotic cell can harbor hundreds to thousands of mtDNA molecules [5] , a number that is tightly regulated [6] and dependent on the cell type and metabolic requirement of the cell [7] . Mutations in mtDNA , including point mutations , rearrangements and deletions , can cause defects in the OXPHOS process . Wild-type ( WT ) and mutant mtDNA can coexist in a cell , a condition known as heteroplasmy , and each cell in a tissue may not have the same composition of mtDNA . Mutant mtDNA molecules have also been shown to accumulate with age in a variety of tissues and organisms [8] , possibly contributing to the general age-related decline in mitochondrial function observed in almost all tissues [9] . However , because of complementation by WT mtDNA , mutant mtDNA molecules need to exceed a threshold between 60–90% before any phenotypic defect manifests [10] . It is therefore important to understand the processes involved in the quality control of mtDNA integrity . Mitochondrial turnover and fusion-fission are key cellular mechanisms involved in the maintenance of mtDNA integrity and mitochondrial function [11] . Mitochondria are continuously turned over by the complementary processes of mitochondrial biogenesis ( mitogenesis ) and mitochondrial autophagy ( mitophagy ) . Mitogenesis is regulated by a nuclear encoded protein family , the peroxisome proliferator-activated receptor gamma coactivator ( PGC ) , e . g . PGC-1α and PGC-1β [12] . The rate of mitogenesis is tightly regulated in response to external stimuli and cellular stress through mitochondrial retrograde signalling [13] . Meanwhile , mitophagy is a selective process , in which mitochondria with lowered mitochondrial membrane potential , an indication of OXPHOS impairment , are preferentially removed [14 , 15] . Together , mitogenesis and mitophagy form the backbone of the mitochondrial quality control process . Perturbations of mitochondrial turnover have been shown to affect mitochondrial function and mtDNA . For example , upregulation of PGC-1α has been shown to lead to higher ATP production in wild-type mice , and delayed onset of myopathy and lifespan extension in a mouse model of mitochondrial myopathy [16] . Similarly , overexpression of PGC-1α in the mtDNA mutator mouse , a transgenic mouse carrying proof-reading deficient mtDNA polymerase γ , improved mitochondrial function despite a slightly higher mtDNA point mutation burden than wild-type mouse [17] . On the other hand , the number of mitochondria with lowered membrane potential increased upon downregulation of autophagy [18] . Meanwhile , rapamycin treatment has been shown to improve mitochondrial function possibly in part through an upregulation of autophagy [19] and increased autophagosomal degradation of mitochondria harboring mtDNA with mutations with a severe OXPHOS deficiency [20] . Mitochondrial fusion and fission are the processes by which two mitochondria combine to form a single organelle and a mitochondrion divides to form two mitochondria , respectively [21] . Together the mitochondrial fusion-fission enables the exchange of mitochondrial components among mitochondria organelles in a cell , including solutes , metabolites and proteins [22 , 23] , and mtDNA [24] . Such exchange provides a buffer against the impact of mtDNA mutations because protein products of WT mtDNA can complement proteins that are missing or defective due to mutations [25] . Disruptions to the machinery of mitochondrial fusion-fission have been shown to affect mitochondria-related function , such as the electron transport chain [26] and apoptosis [27] , and have been associated with neurodegenerative and metabolic diseases [28] . A few experimental studies have also shown that mitochondrial fusion-fission can play a pivotal role in the maintenance of mtDNA integrity . For example , in C . elegans , the removal of helix-distorting mtDNA lesions induced by ultraviolet C ( UVC ) light depends on mitochondrial fusion-fission genes ( fzo-1 , eat-3 , and drp-1 ) [29] . In addition , a severe depletion of either fusion or fission proteins has also been observed to cause a rapid accumulation of deleterious mtDNA mutations and loss of mitochondrial functions in mice and cell culture studies [30 , 31] . A synergistic interaction exists between mitochondrial fusion and mitophagy in the mitochondrial quality control ( QC ) [15 , 29 , 32] . Specifically , depolarized mitochondria with lowered membrane potential have been shown to be less likely to undergo fusion than polarized mitochondria [15] . This selectivity involves the recruitment of Parkin , a cytosolic ubiquitin ligase , to depolarized mitochondria by mitochondrial outer membrane fusion proteins , mitofusins ( Mfn1-2 ) . The binding of Parkin to mitofusins depends on PTEN-induced putative kinase 1 ( PINK1 ) protein . In addition , PINK1 phosphorylates mitofusins , promoting their Parkin-mediated ubiquitination and degradation [33–35] . The selectivity in fusion then leads to the segregation of depolarized mitochondria from the mitochondrial population in the cell . Finally , Parkin localization to damaged mitochondria promotes the targeted removal of these mitochondria through ( selective ) mitophagy [36] . Despite the presence of mitochondrial QC processes , mutant mtDNA molecules often accumulate with age leading to the loss of mitochondrial respiration . The circumstance permitting the expansion of mutant mtDNA is not precisely known , but important in formulating an intervention to the decline in mitochondrial function due to mtDNA mutation accumulation . The focus of this study is to investigate the possible scenario , specifically involving deteriorations in the mitochondrial QC mechanism , that allows the expansion of mtDNA mutations . For this purpose , we would need a comprehensive understanding of how each mitochondrial QC process affects the removal and accumulation of mutant mtDNA . However , a systematic study of the processes involved in mitochondrial QC and their interactions is experimentally challenging . Here , the use of a computational approach through the creation of mathematical models and the simulations and analysis of such models becomes indispensable as an avenue to test hypotheses . For example , we and others have previously simulated mtDNA turnover and the occurrence of clonal expansion of mutant mtDNA due to random segregation of mtDNA populations [37 , 38] . In addition , mathematical models of the fusion-fission process have been previously used to study how the health of mitochondrial populations varies with the rates of mitochondrial turnover and fusion-fission [39 , 40] . In these models , mitochondrial population in a cell has been assumed to be well-mixed . More recently , in separate publications , we and Patel et al . have modeled the fusion-fission among spatially distributed mitochondria organelles in a cell [41 , 42] . We have shown that the ability of mitochondrial retrograde signaling to slow down clonal expansion of mtDNA mutations requires an effective fusion-fission process . In this study , we created a mathematical model of mitochondrial QC mechanism , by extending our previous model of mitochondrial fusion-fission process [42] . We employed a global sensitivity analysis to map out the dependence of mtDNA mutation burden on individual parameters of the mitochondrial QC . Using the corresponding sensitivity coefficients , we could determine the most important process ( es ) which , if perturbed , would significantly affect the ability of the QC to clear mtDNA mutations . Based on the results of model analysis and simulations , we finally propose a scenario that explains the accumulation of mtDNA mutations with age and we suggest potential intervention strategies to counter this accumulation .
Parametric sensitivity analysis ( PSA ) is a systems analysis commonly used for mapping the dependence of biological system behavior on its model parameters [47] . In this analysis , we evaluate sensitivity coefficients , quantifying how much a perturbation in the model parameter affects the behavior of the model . We use the sensitivity coefficients as a relative measure of the importance of parameters , and to rank parameters and the associated biological processes based on their relative impact on system behavior . Parameters with high sensitivity magnitudes correspond to processes that are most critical for the model output . Two types of sensitivity analyses exist , local and global , depending on the nature of the parameter perturbations [47] . A local PSA uses infinitesimal parameter perturbations around well-known parameter value and thus the analysis is typically applied when the uncertainties in the model parameter values are small . Meanwhile , a global sensitivity analysis ( GSA ) involves finite ( large ) parameter perturbations within a range of values , and thus , such analysis is appropriate when the parameter values have relatively large uncertainties . We performed the GSA to elucidate how much R¯MCell depend on the parameters of mitochondrial QC process . More specifically , we calculated the first and second order sensitivity coefficients with respect to each model parameter and every pairwise parameter combination using variance-based global sensitivity coefficients ( see Global Parametric Sensitivity Analysis in Methods ) . The ranges of parameter values are provided in Table 2 . The first order sensitivities Si ( t ) reflect the main effect of the parameter pi on R¯MCell ( t ) . The second order sensitivities Sij ( t ) represent the joint effect of the parameters pi and pj on R¯MCell ( t ) , excluding the main effect of each individual parameter . The second order sensitivity coefficients point to important interactions between two parameters or factors [47] . Here , we used the sensitivity coefficients to rank the parameters according to their effects on R¯MCell , in order to understand the relative impacts of different mitochondrial QC processes on clonal expansion . We calculated the first and second order global sensitivity coefficients of the model as described above , and ranked the parameters and pairs of parameters based on the magnitude of the sensitivities ( the first rank corresponding to the largest magnitude ) . Table 3 provides a list of the important parameter ranking including the average rank over different time points ( for the complete list , see S1 Table ) . The average ranking suggested that the clonal expansion of mtDNA mutations is most highly sensitive to the following parameters ( in decreasing rank ) : replicative advantage ( kR ) , mitophagy selectivity strength ( rD , max ) , fusion selectivity strength ( rfusion , max ) , mitophagy selectivity threshold ( KD ) , and fusion-fission frequency . The existence of replicative advantage ( RA ) of mutant mtDNA , where mutant molecules have higher propensity to replicate than WT ( see Mitochondrial DNA Replication in Methods ) , was the most important determining factor of whether or not a mutant mtDNA molecule would clonally expand . This finding is in agreement with intuition , but the degree of sensitivity is nonetheless insightful . Interestingly , mitochondrial half-life ( related to the parameter kD ) was ranked lower than the selectivity of mitophagy ( see S1 Table ) . Similarly , parameters associated with the selectivity of fusion were ranked higher than the rate of fusion-fission . On the other hand , nuclear retrograde signaling was consistently ranked low , indicating a process of low importance . The low sensitivity of mtDNA mutant expansion with respect to retrograde signaling is perhaps expected as the retrograde response is activated only when mutant mtDNA molecules have already accumulated to a high enough level and thus have little effect on the clearance of new mtDNA mutations . Nevertheless , retrograde signaling could retard the clonal expansion of mtDNA by indirectly reducing stochasticity [38] , an effect that depends in non-trivial ways on fusion-fission rate [42] . Thus , the GSA of our model suggested that among mitochondrial QC processes , mitochondrial fusion-fission , selective mitophagy , and selective fusion are the most relevant targets for preventing the clonal expansion of mtDNA mutations . To gain a deeper understanding of the role of individual mitochondrial QC processes in clearing de novo mutant mtDNA , we simulated the model under different perturbations of model parameters . When investigating the effects of varying fusion-fission frequency , we multiplied the propensities of fusion and fission by the same factor ( >1 to increase or <1 to lower the fusion-fission frequency ) . By doing so , the ratio between the number of fusion and fission events per unit time will stay the same . In the following , we labeled the simulations from different fusion-fission frequencies using the corresponding mixing time τ ( see Fig 2E and Fusion-Fission Parameters in Methods ) . Model simulations using the nominal parameter values in Table 1 correspond to a mixing time of τ = 7 . 5 days . A lower τ indicates more frequent fusion-fission , and vice versa , a higher τ means less frequent fusion-fission . Model simulations as shown in Fig 3A and 3B suggested that the mitochondrial QC can effectively remove a single de novo mtDNA mutation , as indicated by the steady decrease in R¯MCell for τ = 7 . 5 days , even when mutant mtDNA molecules possessed a RA . Here , a zero R¯MCell means that every cell in the population harbors homoplasmic WT mtDNA . In C . elegans , the clearance of UVC-induced mtDNA lesions has been shown to decay following a similar time profile [29] . In the absence of selectivity of mitochondrial fusion , mitochondrial QC could still remove de novo mutant mtDNA without RA ( see Fig 3C ) . Meanwhile , mutant mtDNA with RA was cleared in simulations with τ = 30 days , but not in simulations with τ = 7 . 5 days ( see Fig 3D ) . In general according to the simulations in Fig 3 less frequent fusion-fission has a beneficial effect on the clearance of de novo mtDNA mutations . We observed the opposite effect of fusion-fission frequency on mutant mtDNA clearance when mitophagy selectivity was significantly increased in the absence of fusion selectivity ( see Fig 4A and 4B ) . In this case , more frequent fusion-fission quickened the removal of mutant mtDNA regardless of RA . The reversal of trend can be explained by considering the time scales of mitochondria harboring a high fraction of mutant mtDNA , which we refer to as mutant-rich mitochondria . Mutant-rich mitochondria can arise randomly during fissions , and disappear by fusing with other mitochondria . To illustrate this phenomenon , we simulated the model with only the fusion-fission process in a cell containing a single mutant mtDNA nucleoid ( in the absence of mitochondrial turnover and selectivity of fusion ) . Fig 5 shows the fractional mutation burden RMmito of the mitochondrion harboring the mutant nucleoid , illustrating the appearance and disappearance of mutant-rich mitochondria . When fusion-fission became more frequent ( i . e . for lower τ ) , mutant-rich mitochondria occurred more frequently ( 4 . 1×10–4 day-1mitochondrion-1 for τ = 7 . 5 days vs . 9 . 5x10-5 day-1mitochondrion-1 for τ = 30 days ) . However , with more frequent fusion-fission , these mutant-rich mitochondria had a shorter average lifetime ( 0 . 9 days for τ = 7 . 5 days vs . 3 . 7 days for τ = 30 days ) . Nevertheless , the fraction of time that the mtDNA mutant molecule existed in a mutant-rich mitochondrion did not change , since the numbers of fusion and fission events remained at the same ratio . Therefore , while a quicker fusion-fission process could produce more mutant-rich mitochondria , the shorter existence of these mitochondria meant that there was less time for selective mitophagy to remove them . If the time scale of mitophagy was much longer than the duration of these mutant-rich mitochondria , then mutant mtDNA molecules would get diluted among the population of mitochondria , preventing the clearance of such molecules . In corollary , prolonging the presentation of mutant-rich mitochondria , for example by increasing the selectivity of fusion , should improve the removal of such mitochondria by mitophagy . Simulation results clearly showed that when mutant-rich mitochondria were strongly prohibited from undergoing mitochondrial fusion ( by setting rfusion , max = 100% ) , increasing fusion-fission rate again enhanced the clearance of mtDNA mutations ( see Fig 4C and 4D ) . The faster removal of mutant nucleoids here was a consequence of more frequent generation of mutant-rich mitochondria , their subsequent isolation by selective fusion and removal by selective mitophagy . However , under a milder selectivity of fusion ( rfusion , max = 80% ) , more frequent fusion-fission did not have a significant impact on the removal rate of mutant mtDNA , and even caused a slower clearance of mutant mtDNA when RA was considered as mentioned earlier ( see Fig 3B and 3D ) . This result again demonstrated the importance of prolonging the duration of mutant-rich mitochondria for an effective removal of mutant mtDNA molecules . Taken together , our model simulations illustrate the trade-off in the actions of mitochondrial fusion-fission process in the context of de novo mtDNA mutation removal . On the one hand , the fission process is advantageous in generating mutant-rich mitochondria , which allows phenotypic expression of mtDNA mutations in the corresponding mitochondrion and its preferential removal by mitophagy . On the other hand , the fusion process dilutes mutant mtDNA among the WT population in the cell . While more frequent fusion-fission events increase the occurrence of mutant-rich mitochondria , these mitochondria exist for a shorter period of time . The benefit of faster fusion-fission in removing mutant mtDNA molecules therefore depends on the efficiency by which mitophagy can detect and remove the aforementioned mutant-rich mitochondria . Our model simulations as discussed above showed that the mitochondrial QC can effectively clear mutant mtDNA from the population . However , the fact that mutant mtDNA do clonally expand with ageing suggests the existence of a failure or deterioration in one or more QC mechanisms . Drift in the gene expression level has been observed during ageing [48] . The expression of mitochondrial genes and genes associated with mitochondrial energy production has been observed to drop with age in mice , rats and humans [48] . Moreover , the abundance of mitochondrial proteins can decline by as much as 50% in older individuals [49] . At the same time , genes related to mitochondrial QC , for example PINK1 ( involved in the targeting of depolarized mitochondria for mitophagy ) , are also downregulated in age-related diseases , such as in neurons of Parkinson’s disease patients [50] . A recent mathematical modeling cum experimental study showed that mitochondrial fusion-fission occur less frequently in senescent cells despite of being still tightly coupled [51] . While it is not known if the age-related decline in the associated proteins is the cause of mitochondrial QC deterioration , it is still instructive to understand the implications of a general decline in mitochondrial QC with age . To this end , we performed model simulations where the values of mitophagy selective strength rD , max and fusion selective strength rfusion , max were each lowered by 50% . Under such changes , mitochondrial QC could no longer remove mutant mtDNA molecule with RA ( by comparing Fig 3B with Fig 6A for τ = 7 . 5 and 30 days ) . When the fusion-fission rate remained at the nominal value ( τ = 7 . 5 days ) , mutant mtDNA with RA quickly expanded , as selective mitophagy could not remove mutant-rich mitochondria quickly enough before they disappeared . Interestingly , when the decline in the selectivity was accompanied by a significant drop in the fusion-fission frequency , the mutant molecules could still be cleared , albeit slowly ( see Fig 6A , τ = 120 days ) . Meanwhile , the removal rate of mutant mtDNA without a RA slowed down with lower selectivity of fusion and mitophagy ( compare Fig 3A with Fig 6B ) . Therefore , less frequent fusion-fission could provide a protective effect on mtDNA integrity against declining selectivity of mitochondrial fusion and mitophagy .
Based on the model simulations above and the results of GSA , age-related ( clonal ) expansion of mtDNA mutations likely involves deterioration in either the selectivity of mitophagy or that of mitochondrial fusion or both , i . e . the mechanisms that are responsible in the isolation and targeted removal of mutant mtDNA molecules . Under this condition , slower mitochondrial fusion-fission , which is also expected to happen with age [51 , 52] , may actually retard the accumulation of mutant mtDNA . Our model simulations offered an alternative explanation for the benefit of decelerating fusion-fission with age , without assuming the spread of infectious damage among mitochondria due to fusion-fission [40] . Because of the role of mitochondrial fusion-fission in maintaining normal mitochondrial function , as well as its relevance in mitochondrial diseases , this process has been suggested as a promising target for therapeutic treatment in the context of age-related diseases [53 , 54] . In particular , senescent cells have been observed to suffer from low activity of mitochondrial fusion-fission , which has been suggested as the cause for the accumulation of dysfunctional mitochondria in these cells [52] . The decline in mitochondrial fusion-fission has also been associated with sarcopenia , neurodegenerative disease and the ageing process itself [27 , 55] . Naturally , reversing the age-related change in mitochondrial fusion-fission rate may at first seem reasonable . However , according to the model simulations above , the success of this intervention will sensitively depend on the efficiency of selective mitophagy . An upregulation of mitochondrial fusion-fission in aged cells without a concurrent improvement in the selective mitophagy may aggravate the accumulation of mutant mtDNA molecules and worsen mitochondrial functionality . Any intervention aimed at improving or boosting mitochondrial QC mechanisms will therefore need to consider the balance and interactions among the processes involved , most importantly the selective mitophagy and mitochondrial fusion-fission . Experimental evidence from our laboratory [56] and others [57] has shown modest improvement in mitochondrial function when mitochondrial turnover is increased . However , model simulations and global sensitivity analysis in this study suggest that the selectivity of mitophagy and fusion , not their rates , would be the more interesting targets of slowing down clonal expansion of mtDNA mutations and the corresponding decline of mitochondrial function with age . In summary , simulations of the mathematical model presented in this study illustrated the intricate interplay among the processes involved in the mitochondrial quality control and its implications on the clearance of mutant mtDNA molecules . Our model also produced an interesting insight regarding the dependence of mitochondrial QC on the lifetime of mutant-rich mitochondria that are transiently created by stochastic mixing of mtDNA molecules through mitochondrial fusion-fission . While faster mitochondrial fusion-fission creates more mutant-rich mitochondria , these mitochondria however have shorter lifetimes . Therefore , upregulating mitochondrial fusion-fission is beneficial in removing mutant mtDNA only when mutant-rich mitochondria could be efficiently isolated and removed by the coupled processes of selective mitochondrial fusion and mitophagy . Global sensitivity analysis of the model further confirmed the selectivity of fusion and mitophagy as the most important parameters for clearing mutant mtDNA . Our results thus suggested that the decline of these selective processes is a possible cause of expansion of mutant mtDNA molecules . In this regard , slower fusion-fission could actually lessen the negative effect of such decline on mtDNA integrity .
Here , we consider deleterious mtDNA mutations that cause functional defects in OXPHOS . However , because of mitochondrial heteroplasmy and complementation by WT mtDNA , the energetic impact of the mtDNA mutations will only become apparent when the fraction of mutant mtDNA population exceeds a critical threshold [10] . We choose a sigmoidal function to describe the threshold effect of mtDNA mutations on OXPHOS capacity of a mitochondrion . This choice is supported by an experimental observation showing a sigmoidal relationship between mitochondrial function and mutation load [60] . We formulate the OXPHOS defect function using a standard sigmoidal family of function: s ( RMmito ) = ( RMmito ) mKm+ ( RMmito ) m . The value s ( RMmito ) ∈[0 , 1] reflects the degree of OXPHOS defects , where a value of 0 corresponds to no OXPHOS defect ( at RMmito=0 ) and a value of 1 corresponds to completely dysfunctional OXPHOS . Here , the parameter K represents the midpoint of the threshold effect ( i . e . s ( K ) = 0 . 5 ) , while the parameter m changes the width ( sharpness ) of the threshold region ( see Fig 2B ) . The values of K and m used in this study are provided in Table 1 . Fig 2B shows s ( RMmito ) with parameter K of 75% . The width of s ( RMmito ) is here defined as the range of RMmito corresponding to s = 0 . 1 and s = 0 . 9 . We formulate the propensity of mitophagy of the i-th mitochondrion as the product of the basal mitophagy rate kD and the selectivity function of mitophagy , according to: aD , i ( RM , imito ) =kD ( rD , maxs ( RM , imito ) +1 ) In the model , mitophagy involves removing a single mitochondrion from the cell ( see Fig 2C ) . The parameter kD is related to the half-life of mitochondrial DNA t1/2 , specifically kD = ln ( 2 ) / t1/2 . The selectivity function ( rD , maxs ( RM , imito ) +1 ) becomes larger than 1 for mitochondria with OXPHOS defects , i . e . mitochondria with s ( RM , imito ) > 0 . Thus , the likelihood of dysfunctional mitochondria to be removed by mitophagy will be higher than that of functional mitochondria . The parameter rD , max is related to the degree of mitophagy selectivity , where rD , max + 1 gives the maximum amplification of mitophagy propensity . For example , mitochondria with fully dysfunctional OXPHOS ( with s ( RMmito ) =1 ) are autophagosized at a propensity of ( rD , max + 1 ) times higher than functional mitochondria ( with s ( RMmito ) =0 ) . Note that for mitochondria with s ( RMmito ) =0 , the mitophagy propensity assumes the baseline value aD , i = kD . Under normal conditions , mitochondrial DNA content in a cell is tightly regulated [61] . Defects in OXPHOS due to mtDNA mutations can trigger mitochondrial retrograde signaling that in turn upregulates mitogenesis and mtDNA replication [1] . In mitochondrial myopathies for example , a significant percentage of mtDNA are deleterious and mitochondrial mass and mtDNA copy number have both been shown to increase [62–64] . To reflect this relationship in our model , we calculate the mtDNA nucleoid replication propensity by multiplying the basal propensity of replication aR , 0 with a retrograde response function , such that an increase in the fraction of dysfunctional mitochondria will lead to increased mitochondrial biogenesis . This implementation results in behavior that is in accordance with experimental observation [60] . Overall , we formulate the replication propensity as follows: The number of WT or mutant mtDNA ( W or M , respectively ) is increased by 1 in a nucleoid replication event ( see Fig 2C ) . In the absence of any mutant mtDNA , the nucleation propensity reduces to the basal replication rate aR , 0 . However the presence of mutant mtDNA will trigger the retrograde response , at a degree that varies with the average fraction of mutant mtDNA among mitochondria in the cell . This formulation is motivated by studies of Sarcopenia , where hyperproliferation of mtDNA mutant molecules also leads to OXPHOS dysfunction in muscle fibres despite having a relatively stable WT mtDNA population [62] . Finally , the maximum amplification of nucleoid replication by the retrograde response occurs when s ( R¯Mmito ) =1 , corresponding to a propensity of replication aR ( R¯Mmito ) =kRaR , 0 ( rR , max+1 ) . The parameter kR is used to implement replicative advantage of mutant mtDNA . Many experimental observations reported in the literature suggest the presence of RA of mutant mtDNA , especially for mtDNA with large deletion mutations . For example , in experiments , mtDNA molecules with larger deletions have been shown to be able to re-populate a cell much faster than WT and mutant mtDNA with shorter deletions [65–67] . These observations have subsequently been used to support the hypothesis that mutant mtDNA with large deletions are replicated preferentially or faster than WT mtDNA . While the molecular mechanism of such RA is not precisely known , a recent study has attributed this advantage to a local feedback between the transcription and replication of mtDNA with a deletion mutation [68] . In the model simulations with RA , the parameter kR was set to a value larger than 1 for mutant mtDNA , while kR was set to 1 for WT mtDNA . In simulations without replicative advantage , kR = 1 was used for all mtDNA . In this study , we employed a variance-based global sensitivity analysis to map out the parametric dependence of the model output , which for the purpose of the study , we have set to be the average mutation burden R¯MCell ( t ) . Here , the sensitivities with respect to a parameter or a combination of parameters reflected the variance of R¯MCell ( t ) that was attributed to the ( co- ) variability in the respective parameter ( s ) . More specifically , we calculated the following first and second order sensitivity coefficients [75]: Si ( t ) =V[E ( R¯MCell ( t ) |pi ) ]V ( R¯MCell ( t ) ) Sij ( t ) =V[E ( R¯MCell ( t ) |pi , pj ) ]−V[E ( R¯MCell ( t ) |pi ) ]−V[E ( R¯MCell ( t ) |pj ) ]V ( R¯MCell ( t ) ) where E ( ⋅ ) and V ( ⋅ ) are the expectation and variance functions , respectively . Here , Si ( t ) is the first order sensitivity coefficient of R¯MCell ( t ) with respect to the parameter pi , and Sij ( t ) is the second order sensitivity coefficient of R¯MCell ( t ) with respect to the parameters pi and pj . In the GSA of the mitochondrial QC model , we employed a Latin hypercube sampling to generate 2048 distinct parameter combinations from the parameter ranges given in Table 2 . The parameters K and m associated with the mitophagy selectivity , fusion selectivity and retrograde signaling function , i . e . KD , Kfusion and KR , respectively , were perturbed equally to ensure the same OXPHOS defect threshold in the propensities . We performed model simulations for each parameter combination , generating 2048 trajectories of R¯MCell ( t ) . Finally , we computed Si ( t ) and Sij ( t ) using the GUI-HDMR toolbox in MATLAB [75] . | Mitochondria are responsible for most energy generation in human and animal cells . Loss or pathological alteration of mitochondrial function is a hallmark of many age-related diseases . Mitochondrial dysfunction may be a central and conserved feature of the ageing process . As part of quality control ( QC ) , mitochondria are continually replicated and degraded . Furthermore , two mitochondria can fuse to form a single mitochondrion , and a mitochondrion can divide ( fission ) into two separate organelles . Despite this QC , mutant mitochondrial DNA ( mtDNA ) molecules have been observed to accumulate in cells with age which may lead to mitochondrial dysfunction . In this study , we created a detailed mathematical model of mitochondrial QC and performed model simulations to investigate circumstances allowing or preventing the accumulation of mutant mtDNA . We found that more frequent fusion-fission could quicken mutant mtDNA clearance , but only when mitochondria harboring a high fraction of mutant molecules were strongly prevented from fusing with other mitochondria and selectively degraded . Otherwise , faster fusion-fission would actually enhance the accumulation of mutant mtDNA . Our results suggested that the expansion of mutant mtDNA likely involves a decline in the selectivity of mitochondrial degradation and fusion . This insight might open new avenues for experiment and possible development of future therapies . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [] | 2015 | Context-Dependent Role of Mitochondrial Fusion-Fission in Clonal Expansion of mtDNA Mutations |
There is epidemiological evidence that patients with certain Central Nervous System ( CNS ) disorders have a lower than expected probability of developing some types of Cancer . We tested here the hypothesis that this inverse comorbidity is driven by molecular processes common to CNS disorders and Cancers , and that are deregulated in opposite directions . We conducted transcriptomic meta-analyses of three CNS disorders ( Alzheimer's disease , Parkinson's disease and Schizophrenia ) and three Cancer types ( Lung , Prostate , Colorectal ) previously described with inverse comorbidities . A significant overlap was observed between the genes upregulated in CNS disorders and downregulated in Cancers , as well as between the genes downregulated in CNS disorders and upregulated in Cancers . We also observed expression deregulations in opposite directions at the level of pathways . Our analysis points to specific genes and pathways , the upregulation of which could increase the incidence of CNS disorders and simultaneously lower the risk of developing Cancer , while the downregulation of another set of genes and pathways could contribute to a decrease in the incidence of CNS disorders while increasing the Cancer risk . These results reinforce the previously proposed involvement of the PIN1 gene , Wnt and P53 pathways , and reveal potential new candidates , in particular related with protein degradation processes .
Epidemiological evidences point to a lower-than-expected probability of developing some types of Cancer in certain CNS disorders , including Alzheimer's disease ( AD ) , Parkinson's disease ( PD ) and Schizophrenia ( SCZ ) [1]–[6] . Our current understanding of such inverse comorbidities suggests that this phenomenon is influenced by environmental factors , drug treatments and other aspects related with disease diagnosis . Genetics can additionally contribute to the inverse comorbidity between complex diseases , together with these external factors ( for review , see [3]–[7] ) . In particular , we propose the deregulation in opposite directions of a common set of genes and pathways as an underlying cause of inverse comorbidities . To investigate the biological plausibility of this hypothesis , a basic initial step is to establish the existence of inverse gene expression deregulations ( i . e . , down- versus up-regulations ) in CNS disorders and Cancers . Towards this objective , we have performed integrative meta-analyses of collections of gene expression data , publically available for AD , PD and SCZ , and Lung ( LC ) , Colorectal ( CRC ) and Prostate ( PC ) Cancers . Clinical and epidemiological data previously reported inverse comorbidities for these complex disorders , according to population studies assessing the Cancer risks among patients with CNS disorders [8]–[17] .
For each CNS disorder and Cancer type independently , we undertook meta-analyses from a large collection of microarray gene expression datasets to identify the genes that are significantly up- and down-regulated in disease when compared with their corresponding healthy control samples ( Differentially Expressed Genes – DEGs – , FDR corrected p-value ( q-value ) <0 . 05 , see Methods and Table S1 ) . Then , the DEGs of the CNS disorders and Cancer types were compared to each others . There were significant overlaps ( Fisher's exact test , corrected p-value ( q-value ) <0 . 05 , see Methods ) between the DEGs upregulated in CNS disorders and those downregulated in Cancers . Similarly , DEGs downregulated in CNS disorders overlapped significantly with DEGs upregulated in Cancers ( Figure 1A ) . Significant overlaps between DEGs deregulated in opposite directions in CNS disorders and Cancers are still observed while setting more stringent cutoffs for the detection of DEGs ( qvalues lower than 0 . 005 , 0 . 0005 , 0 . 00005 and 0 . 000005 , ) . A significant overlap between DEGs deregulated in the same direction was only identified in the case of CRC and PD upregulated genes ( Figure 1A ) . A molecular interpretation of the inverse comorbidity between CNS disorders and Cancers could be that the downregulation of certain genes would at the same time increase the risk of developing CNS disorders , while reducing the risk of developing Cancers . The upregulation of other genes would reduce the risk of developing CNS disorders and increase the risk of developing Cancers . We then compared the CNS disorder and Cancer DEGs with DEGs of a number of diseases for which , to our knowledge , inverse comorbidities have not been reported in the literature . These diseases , for which large enough expression datasets were available , included Asthma , HIV , Malaria , Dystrophy and Sarcoidosis ( see Methods ) . Significant overlaps were observed between DEGs of all these diseases and DEGS of CNS disorders or Cancers ( Figure 1B ) . However , patterns of expression deregulation in opposite directions , which were found to be characteristic of the relation between CNS disorders and Cancers , are in most cases not observed with these other genetic or infectious diseases ( Figure 1B ) . Indeed , the overlaps are predominantly significant between DEGs deregulated in the same directions , i . e . between upregulated genes of the different diseases ( or conversely between down-regulated genes ) , and could be a signature of putative positive comorbidities . It is to note that Malaria and CNS disorders DEGs present overlaps between DEGs deregulated in opposite directions , contrarily to what is detected for other diseases . This observation will require additional research . Overall , these observations support the indication of a signature for inverse comorbidity in gene expression deregulations in opposite directions . The PIN1 gene has been proposed previously as a putative link between the pathogeneses of AD and Cancer [4] . Through the isomerization of a proline preceded by phosphorylated Ser/Thr residues , the PIN1 protein is known to be a key regulator of cell division [18] . PIN1 gene is typically overexpressed in human Cancers and as such , it has been assessed as a potential target for anticancer drugs [4] . In addition , PIN1 is depleted in AD , it has been shown to restore the function of the phosphorylated tau protein , and mouse models in which this protein is knocked-down present neurodegenerative phenotypes [18]–[19] . Our transcriptomic meta-analyses confirm and extend these observations as the expression of PIN1 is downregulated in AD and PD , and upregulated in CRC ( Table S2 ) . Another interesting case is the ATP13A2 gene , involved in the intracellular cation homeostasis . ATP13A2 is part of a list established by Devine et al . of familial PD genes frequently mutated in Cancers [5] . Indeed , loss-of-function mutations of ATP13A2 have been associated with early-onset Parkinsonism , and somatic mutations have been independently observed in Cancer [5] . We identified ATP13A2 as downregulated in AD and PD , and upregulated in the three Cancer types considered ( Table S2 ) . In the light of these findings , our approach appears to be capable of identifying candidate genes potentially associated with inverse comorbidity . In particular , 74 genes may be of interest since they are simultaneously downregulated in the three CNS disorders and upregulated in the three Cancer types examined ( Table 1 ) . RNA splicing ( four genes: PPIH , LSM4 , NUDT21 , SRSF2 ) and aminoacyl t-RNA ligases ( three genes: FARSA , IARS , IARS2 ) represent particularly interesting functions . We also pinpoint two genes involved in lipid biogenesis ( ACLY and MECR ) , and other two are transcription factors: NME2 and TFCP2 , for which a genetic association with AD is debated [20] . Finally , two other genes , OAZ2 and the spermine synthase SMS , are dedicated to polyamine metabolic processes . Interestingly , defects in the spermine synthase gene are associated with the X-linked mental retardation Snyder-Robison syndrom [21] , and spermine is often the most abundant polyamine in Cancers [22] . The polyamine metabolic process hence may play a role in the pathological mechanisms of both CNS disorders and Cancers . Conversely , 19 genes are simultaneously upregulated in the three CNS disorders and downregulated in the three Cancer types examined ( Table 2 ) , including for instance six genes involved in signal transduction ( TNFRSF1A , CDKN1A , NFKBIA , PTH1R , IL4R , MID1 ) . Particularly , NFKBIA is an interesting candidate because this gene is often deleted in glioblastoma [23] , although to our knowledge no mutations or polymorphisms have been described in CNS disorders . In order to enhance the functional interpretation of the molecular bases of inverse comorbidity , we broaden the comparisons of expression deregulations by considering pathways instead of individual genes [24] . We identified the pathways that were significantly up- and downregulated ( GSEA analyses , q-value<0 . 05 , see Methods and Table S3 ) in each of the six diseases independently . Among all the KEGG [25] pathways significantly up- and down-regulated in the 6 diseases , 30 are shared by CNS disorders and Cancers ( i . e . , significantly deregulated in at least 1 CNS disorder and 1 Cancer type ) . Strikingly , of these 30 shared pathways , 24 ( 80% ) are deregulated in opposite directions in CNS disorders and Cancers ( Figure 2 , 63% and 86% for the Biocarta ( http://www . biocarta . com/ ) and Reactome [26] databases , respectively , Figure S2 ) . The p53 signalling pathway is an anticipated candidate for deregulations in these diseases and for a role in inverse comorbidity [4] . Indeed , deregulations of the p53 signalling pathway are associated with the initiation and progression of Cancers , while recent studies also point to a role for this pathway in CNS disorders [27] . As such , specific polymorphisms in the TP53 gene are found in SCZ patients [27] . Although the TP53 gene itself does not appear to be differentially regulated in our analysis , the p53 pathway is upregulated in CRC and LC , while it is downregulated in PD , AD and SCZ ( Reactome database; Figure S2 , Table S3 ) . Similarly , the Wnt pathway may be particularly relevant as mutations in the genes encoding APC and β-catenin , elements of the Wnt pathway , have been described in CRC , while β-amyloid induced neurotoxicity in AD has been associated with impaired Wnt signalling [4] , [18] . Furthermore , alterations in the Wnt signalling pathway are known to be involved in SCZ [28] . In our meta-analyses , we found the Wnt pathway to be downregulated in AD and PD , and upregulated in CRC ( Reactome database; Figure S2 ) . Aside the Wnt and p53 pathways , our analysis reveals other pathways related to protein folding and protein degradation displaying patterns of downregulation in CNS disorders and upregulation in Cancers , and that may be relevant for inverse comorbidity . For instance , the Ubiquitin/Proteasome system is consistently downregulated in CNS disorders and upregulated in Cancers according to the three pathway databases analyzed ( Figure 2 , Figure S2 , Table S3 ) . The inverse relationship between the levels of expression deregulations of these pathways possibly suggests opposite roles in CNS disorders and Cancers . A detailed examination of the KEGG pathways deregulated in opposite directions in CNS disorders and Cancers finally revealed that 89% of the KEGG pathways that were upregulated in Cancers and downregulated in CNS disorders are related to Metabolism and Genetic Information Processing ( Figure 2 , Figure 3 ) . By contrast , the pathways downregulated in Cancers and upregulated in CNS disorders are related to the cell's communication with its environment ( Environmental Information Processing and Organismal System; Figure 2 , Figure 3 ) . Hence , global regulations of cellular activity may account for a protective effect between inversely comorbid diseases . Further analyses will be necessary to conclude to a direct protective effect of gene expression deregulations in cancer-prone tissues of patients suffering from CNS disorders . Indeed , the DEGs analyzed in this study are computed for each disease in the corresponding affected tissues , and cannot be extrapolated to gene expression deregulations in other tissues of the same patients . However , despite these limitations , the identification of antagonistically deregulated genes and pathways in complex diseases that have been previously described as inversely comorbid provides , to our knowledge , the first systematic insights into the possible molecular basis of these associations . It suggests that the upregulation of a set of genes or processes could increase the incidence of CNS disorders and simultaneously lower the chances of developing Cancers , while the downregulation of another set of genes or processes could contribute to a decrease in the incidence of CNS disorders while increasing the Cancer risks . The individuals delivering post-mortem brain samples in CNS disorders , or tumor tissues in the case of Cancers , are likely to have received drug treatments . Hence , the observed expression deregulations could be the consequence of the drugs administered to the patients . If this is the case , it can be hypothesized that some of the drugs used to treat CNS disorders might be able to revert the expression of a number of Cancer genes . In this context , the repurposing of drugs from the CNS to the Cancer field could open new therapeutic avenues . Indeed some punctual observations have been made . For example , the thioridazine , an anti-psychotic drug antagonizing the dopamine receptor and potentially able to alter physiological states and expression patterns , have been reported to target cancer stem cells selectively [29] . Finally , the analyses of inverse expression deregulations could serve as a new approach to investigate possible relations between complex diseases , of which the ones reported here between CNS disorders and Cancers can be considered as an initial example .
Gene expression raw data ( CEL files ) were downloaded from NCBI GEO omnibus ( GEO , http://www . ncbi . nlm . nih . gov/geo/ ) , EBI ArrayExpress ( AE , http://www . ebi . ac . uk/arrayexpress/ ) and Stanley Medical Research Institute , Online Genomics Database ( SMRI , https://www . stanleygenomics . org ) for Colorectal ( CRC ) , Lung ( LC ) and Prostate ( PC ) Cancers , Alzheimer's disease ( AD ) , Parkinson's disease ( PD ) and Schizophrenia ( SZC ) , and for Asthma , HIV , Malaria , Dystrophy , Sarcoidosis ( Text S1 ) . For each disease , studies were filtered to select only the ones profiling at least 9 samples for disease and control cases , with Affymetrix arrays ( GeneChip Human Genome U133 Plus 2 . 0 , GeneChip Human Genome U133A and GeneChip Human Genome U133A 2 . 0 containing 23 , 945 , 14 , 538 and 14 , 538 genes , respectively ) . For CNS disorders , only studies that measure gene expression in brain tissues were selected . For Cancers , only gene expression studies carried out in the LC , CRC and PC tumor tissues were considered . The collected microarray data from the different studies were normalized with frozen Robust Multiarray Analysis ( fRMA ) [30] from the R Affy package [31] . Then , microarray meta-analyses were undertaken for each disease independently using the R MetaDE package [32] . MetaDE implements meta-analysis methods for differential expression analysis , and we used the Fixed Effects Model ( FEM ) [33] . This model assumes that the standardized effect sizes can be combined between the different studies , and that the variations in observed effects are only due to random error [34]–[35] . Similar results were obtained with the Random Effects Model ( REM ) approach that allows heterogeneity in the effect sizes between the different datasets ( unpublished observations ) . The meta-analyses led to the identification of genes up- and down-regulated in each disease , and significant differentially expressed genes ( DEGs ) were selected as those displaying a FDR corrected p-value ( q-value ) <0 . 05 . Four other q-value cutoffs ( 0 . 005 , 0 . 0005 , 0 . 00005 and 0 . 000005 ) were selected to validate our results on more stringent DEGs sets ( Figure S1 ) . Each CNS disorder DEGs were compared to each Cancer type DEGs , and the significances of the overlaps between the DEGs were assessed by a one-tailed Fisher's exact test , corrected for multiple testing by the Bonferroni approach ( Figure 1A , Figure S1 ) . The background number of genes necessary for the Fisher's test was set to 14 , 538 . The same procedure was applied for Cancers , CNS disorders and Asthma , HIV , Malaria , Dystrophy and Sarcoidosis ( Figure 1B ) . For each CNS disorder and Cancer type independently , a gene set enrichment analysis was undertaken using GSEA [36] on the output of the meta-analyses , and focusing on KEGG [24] , Biocarta ( http://www . biocarta . com/ ) and Reactome [26] pathway databases . Significant pathways were selected as those with q-value ( FDR ) <0 . 05 . Significant pathways in each disease were then compared to each others , and a network of pathways was built ( Figure 2 , Figure S2 ) . For the KEGG pathways , further classification of the pathways in Metabolism , Genetic Information Processing , Cellular Processes , Environmental Processes and Organismal Processes , as provided by KEGG , was done ( Figure 2 , Table S2 ) . Pathways corresponding to Human Diseases were discarded . | A lower-than-expected probability of developing certain types of Cancer has been observed in patients with CNS disorders , including Alzheimer's disease , Parkinson's disease or Schizophrenia . Understanding such a protective effect could be the key to finding novel treatments for both types of conditions , for instance thanks to drug repurposing . However , little is known about the underlying mechanisms for these intriguing inverse comorbidities . Although environmental causes , drug treatments or lower screening surveys might contribute to the inverse comorbidity between complex disorders , we propose that inverse comorbidity is , at least in part , due to genetic factors . We observe here that a common set of genes and biological processes are deregulated in opposite directions in CNS disorders and Cancers , i . e . upregulated in CNS disorders and downregulated in Cancers , or vice versa . We propose the alluring hypothesis that the deregulation of these genes and processes could promote CNS disorders and simultaneously lower the initiation or progression of Cancers . | [
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Methods"
] | [
"cancer",
"genetics",
"gene",
"expression",
"genetics",
"biology",
"computational",
"biology",
"gene",
"networks"
] | 2014 | Molecular Evidence for the Inverse Comorbidity between Central Nervous System Disorders and Cancers Detected by Transcriptomic Meta-analyses |
Deregulation of the Wnt signal transduction pathway underlies numerous congenital disorders and cancers . Axin , a concentration-limiting scaffold protein , facilitates assembly of a “destruction complex” that prevents signaling in the unstimulated state and a plasma membrane-associated “signalosome” that activates signaling following Wnt stimulation . In the classical model , Axin is cytoplasmic under basal conditions , but relocates to the cell membrane after Wnt exposure; however , due to the very low levels of endogenous Axin , this model is based largely on examination of Axin at supraphysiological levels . Here , we analyze the subcellular distribution of endogenous Drosophila Axin in vivo and find that a pool of Axin localizes to cell membrane proximal puncta even in the absence of Wnt stimulation . Axin localization in these puncta is dependent on the destruction complex component Adenomatous polyposis coli ( Apc ) . In the unstimulated state , the membrane association of Axin increases its Tankyrase-dependent ADP-ribosylation and consequent proteasomal degradation to control its basal levels . Furthermore , Wnt stimulation does not result in a bulk redistribution of Axin from cytoplasmic to membrane pools , but causes an initial increase of Axin in both of these pools , with concomitant changes in two post-translational modifications , followed by Axin proteolysis hours later . Finally , the ADP-ribosylated Axin that increases rapidly following Wnt stimulation is membrane associated . We conclude that even in the unstimulated state , a pool of Axin forms membrane-proximal puncta that are dependent on Apc , and that membrane association regulates both Axin levels and Axin’s role in the rapid activation of signaling that follows Wnt exposure .
The Wnt/Wingless signal transduction pathway directs fundamental cellular processes during animal development and tissue homeostasis , whereas Wnt pathway deregulation results in numerous cancers and congenital disorders [1 , 2] . In the unstimulated state , the concentration-limiting scaffold protein Axin facilitates assembly of a cytoplasmic “destruction complex” that includes the tumor suppressor Adenomatous polyposis coli ( Apc ) as well as glycogen synthase kinase 3 ( GSK3 ) and targets the transcriptional activator β-catenin for proteasomal degradation . Binding of Wnt ligands to their transmembrane co-receptors LRP6/Arrow and Frizzled induces rapid phosphorylation of the intracellular tail of LRP6 , creating binding sites for Axin [3–6] . The consequent recruitment of Axin , GSK3 , and the cytoplasmic component Dishevelled to LRP6 and Frizzled promotes assembly of an activation complex termed the “signalosome” [7] . Axin is thought to facilitate signaling by acting as a scaffold for the signalosome [3 , 5] and by promoting LRP6 phosphorylation following Wnt stimulation [4] , although the initial phosphorylation of LRP6 may occur independently of Axin [8] . Signalosome assembly results in β-catenin stabilization and the transcriptional regulation of Wnt pathway target genes [3 , 4 , 9] . The levels of Axin under basal conditions are very low [10 , 11] , and regulated by the ADP-ribose polymerase Tankyrase ( Tnks ) . Tnks-mediated ADP-ribosylation targets Axin for ubiquitin-dependent proteasomal degradation [12–14] . The role of Tnks in controlling Axin levels is conserved in Drosophila [15–18] . Due to functional redundancy in vertebrate Tnks homologs [19] , the in vivo settings in which mammalian Tnks promotes Wnt signaling remain uncertain , but it is known that in Drosophila , the requirement for Tnks is context-specific , as Tnks is dispensable for many Wingless-dependent developmental processes [15 , 17] . However , in the adult Drosophila intestine , Tnks is essential for target gene activation within regions where Wingless is present at relatively low concentration and promotes the Wingless-dependent regulation of midgut stem cell proliferation [17 , 20] . Furthermore , when endogenous Axin levels are increased by only two-fold , Tnks is required for Wingless-dependent cell fate specification in the embryonic epidermis [18] . In addition , Tnks-mediated ADP-ribosylation of Axin promotes not only Axin proteolysis in the unstimulated state , but also the rapid transition in Axin activity that follows Wnt stimulation and the interaction of Axin with phospho-LRP6 , which is a key step in the activation of signaling [18] . In the classical model , Axin localizes in the cytoplasm under basal conditions , but relocates to the plasma membrane following Wnt stimulation [7 , 21] . However analysis of Axin regulation under physiological conditions has been impeded by the very low levels of endogenous Axin . Therefore , previous in vivo work regarding the relocation of Axin that follows Wnt stimulation was based largely on overexpression of Axin to levels that completely inhibited Wnt signaling in Drosophila embryos , thus disrupting its physiological regulation [21 , 22] . In these studies , overexpressed Axin was described as “dots” that localized either throughout the cytoplasm in the unstimulated state or primarily at the plasma membrane after Wingless stimulation . Based on these findings , the authors concluded that Wingless exposure induces the bulk relocation of Axin from cytoplasm to plasma membrane . However , the Axin-GFP fusion protein used in these studies was not only highly overexpressed but also aberrantly stabilized; this Axin-GFP abrogated Wnt signaling and was refractory to the degradation of Axin that occurs several hours after Wnt stimulation in Drosophila [18 , 23] and vertebrate cells [8 , 24–26] . Here , we investigate the regulation of endogenous Drosophila Axin in vivo . We find that even in the unstimulated state , a pool of Axin is localized in puncta at the cell periphery through association with the plasma membrane and/or vesicles just proximal to the membrane . Axin’s localization at these peripheral puncta is independent of Wingless stimulation , but dependent on Apc . Furthermore , membrane association increases the ADP-ribosylation of Axin and thereby promotes Tnks-mediated proteasomal degradation under basal conditions . Moreover , we find no evidence for the bulk redistribution of Axin from cytosolic to membrane-associated pools following Wingless stimulation in vivo or in cultured embryonic cells . Instead , we find that Wingless exposure initially results in increased levels of both cytoplasmic and membrane-associated Axin accompanied by changes in two post-translational modifications , followed by Axin proteolysis hours later . Importantly , Wingless stimulation induces a preferential increase in the pool of ADP-ribosylated Axin associated with the membrane . As ADP-ribosylation promotes the interaction of Axin with phospho-LRP6 [18] , the increased levels of membrane-associated ADP-ribosylated Axin may facilitate rapid and robust Wnt pathway activation through both local enrichment at the membrane and increased interaction with LRP6 . We conclude that the formation of membrane-proximal Axin puncta , some of which may represent the sites of the destruction complex , depends on Apc , and that membrane association both regulates Axin levels and leaves Axin poised for rapid activation of the Wnt pathway .
To characterize the subcellular localization of endogenous Axin , we examined confocal sections of larval imaginal disc epithelia stained with an Axin antibody at subapical ( Fig 1A–1C ) and basolateral ( Fig 1D–1I ) levels . We confirmed the specificity of the Axin antibody by comparing the Axin signal in wild-type cells with that in juxtaposed mitotic clones of Axin null mutant cells ( labeled -/- in panel ) . Double labeling with Armadillo/β-catenin antibody provided a reference for subcellular localization . As expected , Armadillo was enriched at the adherens junctions that demarcate the subapical plasma membrane [27] ( Fig 1C ) and was also observed in the cytoplasm of Axin null mutant cells , resulting from inactivation of the destruction complex [28] ( Fig 1A and 1C ) . Weak background staining was observed in the Axin null mutant clones; however the Axin antibody revealed strong and uniform endogenous Axin signal in the apical cytoplasm of wild-type cells , at and above the level of the adherens junctions ( Fig 1B ) . In contrast with its uniform localization in apical sections , Axin signal was prominent at the cell cortex in basolateral sections , and in particular at the vertices between neighboring cells ( Fig 1D–1F ) . The Axin signal was markedly reduced in Axin null mutant cells , verifying its specificity . To determine the localization of cortical Axin with respect to the basolateral cell membrane , we co-stained imaginal discs with antibodies against Axin and Fas III , which demarcates the basolateral cell membrane [29] . Axin staining partially overlapped that of Fas III , but was not identical ( Fig 1G–1I ) : close examination revealed Axin in puncta at or juxtaposed with the basolateral plasma membrane ( Fig 1G’–1I’ ) , even at regions far from the Wingless-expressing cells . To determine whether this unanticipated Axin localization was restricted to larval stages , we also co-stained pupal wings , in which Wingless expression in very restricted [30] , with antibodies against Axin and Discs Large ( Dlg ) , another basolateral membrane marker [31] . Importantly , by comparison with cells in the larval wing disc , the larger size of cells in pupal wings permitted unequivocal discrimination between cell membrane and cytoplasm . As observed in larval wing discs , Axin was present ubiquitously in puncta juxtaposed with the basolateral membrane of epithelial cells in pupal wings ( Fig 1J–1L ) . These findings revealed that in addition to a diffusely distributed cytosolic pool , Axin is localized in membrane-proximal puncta throughout the wing epithelium at multiple developmental stages in the absence of Wingless stimulation , indicating that the localization of Axin to membrane-associated puncta occurs independently of Wingless exposure . To distinguish the cytoplasmic and membrane-associated Axin pools using an independent approach , we fractionated lysates from cultured Drosophila embryonic S2R+ cells and analyzed these lysates by immunoblotting with Axin antibody . The transmembrane protein LRP6/Arrow and the cytosolic protein α-tubulin were used as controls for the efficiency of subcellular fractionation ( Fig 1M ) . Consistent with our in vivo observations , endogenous Axin was present not only in the cytosolic fraction , but also the membrane fraction , even in the absence of Wingless stimulation ( Fig 1M ) . Quantification indicated that endogenous Axin is present almost equally in the cytoplasmic and membrane fractions ( Fig 1N ) . Notably , cytoplasmic and membrane-associated Axin displayed different migration rates in SDS-PAGE , suggesting that Axin from each pool is subject to differential post-translational modification in these cells ( Fig 1M ) . To further investigate the subcellular distribution of endogenous Axin in the unstimulated state , we fractionated lysates from wild-type Drosophila embryos that were collected within 2 . 5 hours of development , which is prior to the onset of Wingless expression [32] . Immunoblots with Axin antibody revealed that Axin was present nearly equally in the cytosolic and membrane fractions ( Fig 1O and 1P ) . Although we cannot exclude the possibility that some Axin is present in cytoplasmic vesicles , these results , coupled with our immunostaining data from larval and pupal wings , support a model in which a pool of Axin is juxtaposed with the membrane in the unstimulated state . We sought to determine if the localization of Axin to membrane-associated puncta occurs through interaction with its binding partners . We first examined cells that were devoid of the co-receptors required for the response to Wingless stimulation: LRP6/Arrow and the functionally redundant Frizzled ( Fz ) and Frizzled 2 ( Dfz2 ) [33–36] . In clones of arrow null mutant cells in the larval wing discs , the localization of Axin near the basolateral plasma membrane was the same as in neighboring wild-type cells ( S1A–S1C Fig ) . Similarly , in clones of fz Dfz2 double null mutant cells , the localization of Axin was indistinguishable from that of the neighboring wild-type cells ( S1D–S1F Fig ) . These findings suggested that Axin localization to peripheral puncta does not require interaction with Frizzled or Arrow , and provided further evidence that the existence of a membrane-associated Axin pool does not require Wingless pathway activation nor the interaction of Wingless with Arrow or Frizzled . Apc2 , which binds Axin in the destruction complex , is present both in the cytoplasm and at the cell cortex in Drosophila embryos [37 , 38] . To determine the subcellular localization of Apc2 in larval wing imaginal discs , we examined endogenous Apc2 using an Apc2 antibody . Immunostaining revealed strong signal in wild-type tissue that was markedly reduced in juxtaposed clones of Apc2 null mutant cells , verifying the specificity of the Apc2 antibody ( Fig 2A–2C ) . Double staining with Fas III antibody revealed that at the level of the basolateral cell membrane , Apc2 was present not only in the cytoplasm , but also enriched at the cell cortex ( Fig 2E ) . The Apc2 signal overlapped , but was distinct from Fas III and was prominent at the vertices between neighboring cells ( Fig 2D–2F ) . Double labeling experiments revealed that Apc2 and Axin co-localized at some puncta that were juxtaposed with cell membrane ( Fig 2G–2I ) , suggesting that these puncta may represent sites of the destruction complex . As Axin partially co-localizes with Apc2 , we sought to determine if Apc promotes Axin localization in puncta juxtaposed with the cell membrane . We examined Axin staining in either Apc2 or Apc1 null mutant clones in larval wing imaginal discs . In sharp contrast to its punctate staining at the periphery of wild-type cells , Axin staining was more diffuse in Apc2 or Apc1 mutant clones , although some Axin puncta remained ( Fig 2J–2N and 2O–2S ) . Fas III staining confirmed that the morphology of cells in Apc2 or Apc1 clones was the same as the neighboring wild-type cells ( Fig 2M and 2R ) , indicating that the decreased Axin localization to puncta at the cell periphery was not a secondary consequence of disrupted cell morphology . Importantly , previous genetic and biochemical studies had demonstrated that endogenous Axin is not destabilized as a consequence of Apc loss [16]; therefore , the decreased localization of Axin in peripheral puncta is not secondary to decreased Axin levels in Apc mutant cells . Based on these results , we conclude that Apc promotes Axin localization to puncta at the cell periphery in the unstimulated state . Next , we sought to analyze the function and regulation of membrane-associated Axin . To address this , we generated an Axin transgene ( Myr-Axin-V5 ) in which we added to the Axin amino-terminus a ten amino acid myristoylation sequence from the Drosophila Src protein [39] , which is known to target proteins to membranes [40 , 41] . Myr-Axin-V5 and an Axin-V5 transgene that does not contain the myristoylation sequence but is otherwise identical [18] , were integrated at the same genomic site to allow for their direct comparison in the absence of transcriptional position effects . To investigate the effect of the myristoylation sequence on the subcellular localization of Axin , we analyzed the adult midgut , in which the large size of absorptive epithelial cells ( enterocytes ) permits unequivocal distinction between plasma membrane and cytoplasm . In contrast with Axin-V5 , which localized to both cytoplasm and cell membrane , Myr-Axin-V5 was localized predominantly at the cell membrane ( Fig 3A and 3B ) , indicating that the myristoylation sequence was effective in targeting Axin to the plasma membrane , consistent with previous work [42] . To test this conclusion using an independent approach , we expressed Myr-Axin-V5 or Axin-V5 in S2R+ cells and determined their distribution using subcellular fractionation . Axin-V5 was distributed nearly equally in the cytoplasmic and membrane fractions; in contrast , Myr-Axin-V5 was highly enriched in the membrane fraction ( Fig 3C ) . We conclude that the myristoylation sequence is sufficient for the membrane targeting of Axin . We previously established a system to express an Axin-V5 transgene within the Axin physiological threshold in larval wing imaginal discs using the C765-Gal4 or 71B-Gal4 drivers [16] . We sought to use this system to express membrane-targeted Axin within physiological range and to study its regulation . Therefore , we first examined Axin protein levels in lysates from control wing discs or those expressing either the Myr-Axin-V5 or Axin-V5 transgene with the C765-Gal4 driver ( Fig 3D ) . Quantification revealed a two-fold and four-fold increase in Axin levels in wing discs expressing Myr-Axin-V5 or Axin-V5 respectively , by comparison with controls expressing only endogenous Axin ( Fig 3E ) . These levels of Myr-Axin-V5 and Axin-V5 are below the threshold at which Axin overexpression disrupts Wingless signaling in wing discs [16] . To test this conclusion using an independent approach , we determined whether Myr-Axin-V5 expression at these levels disrupted Wingless signaling in wing discs . Staining with an antibody against Senseless , a Wingless signaling readout at the dorso-ventral boundary of the third instar larval wing imaginal disc , revealed a pattern that was indistinguishable from controls ( Fig 3F–3H ) . In addition , expression of Axin-V5 or Myr-Axin-V5 using the C765-Gal4 driver did not disrupt the morphology of adult wings ( Fig 3I–3K and 3I’–3K’ ) . Similar results were obtained when Axin-V5 or Myr-Axin-V5 was expressed using the 71B-Gal4 driver ( S2 Fig ) . Taken together , these findings indicated that when expressed within physiological range , membrane-tethered Axin does not disrupt normal Wingless signaling . To determine whether membrane-targeted Axin expressed within physiological range was sufficient to replace the function of endogenous Axin , we tested whether Myr-Axin-V5 could restore normal signaling in Axin null mutants . As expected , senseless was ectopically expressed in Axin null mutant clones , resulting from the aberrant activation of Wingless signaling ( Fig 4A–4C ) . In contrast , expression of Myr-Axin-V5 in wing discs using the 71B-Gal4 driver fully prevented ectopic senseless expression even in the absence of endogenous Axin , and importantly , did not disrupt physiological senseless expression ( Fig 4D–4F ) . Similarly , the aberrant stabilization of cytoplasmic Armadillo in Axin mutant cells ( Fig 4G–4I ) was fully prevented in cells expressing Myr-Axin-V5 ( Fig 4J–4L ) . These results suggested that membrane-bound Axin can functionally replace endogenous Axin in both the destruction complex and signalosome . Our results in wing imaginal discs are consistent with the ability of membrane-bound Axin to functionally replace endogenous Axin during embryogenesis [42] . Taken together , these findings suggested that membrane-associated Axin is sufficient for proper regulation of Wingless signaling . We observed , unexpectedly , that levels of Myr-Axin-V5 were lower than those of Axin-V5 , as revealed by immunoblotting of lysates from cultured cells and larval wing discs expressing these respective proteins ( Fig 3C , lane 1 and lane 4 and Fig 3D and 3E ) , suggesting that the membrane targeting of Axin may result in its destabilization . To rule out the possibility that the addition of amino acids to the Axin amino-terminus inadvertently disrupted Axin regulation , we generated a control transgene , MyrG-A-Axin-V5 , in which substitution of a single amino acid ( Gly to Ala ) within the myristoylation sequence is known to abolish myristoylation and membrane localization [43] . We found that MyrG-A-Axin-V5 was present at much higher levels than Myr-Axin-V5 in larval wing discs ( Fig 5A and 5B ) . The same result was observed in lysates from adult midguts ( S3A Fig ) , suggesting that the decreased levels of Myr-Axin-V5 were a specific consequence of its membrane association . Together , these findings suggested that the association of Axin with the membrane promotes its proteolysis . As we had found that targeting Axin to the membrane results in its destabilization , we sought to determine whether Tnks , which is known to target Axin for degradation , promotes the proteolysis of membrane-associated Axin . To address this question , we generated a transgene encoding membrane-tethered Axin lacking the Tnks binding domain ( Myr-AxinΔTBD-V5 ) . We postulated that if Tnks were important for Axin degradation at the membrane , then Myr-AxinΔTBD-V5 would be more stable than Myr-Axin-V5 . Expression of Myr-Axin-V5 or Myr-AxinΔTBD-V5 in S2R+ cells supported this hypothesis; Myr-AxinΔTBD-V5 was present at higher levels than Myr-Axin-V5 ( Fig 5C ) . These results suggested that Axin degradation at the membrane required the Tnks binding domain of Axin . To test whether Tnks promotes the degradation of membrane-associated Axin in vivo , we expressed Axin-V5 or Myr-Axin-V5 using the C765-Gal4 driver in wing discs of either wild-type or Tnks mutants and compared Axin levels . We found that inactivation of Tnks significantly increased the levels of Myr-Axin-V5 to levels comparable to Axin-V5 ( Fig 5D and 5E ) , suggesting that Tnks-dependent proteolysis is a major mechanism that promotes the degradation of membrane-associated Axin . Supporting this conclusion , elimination of Tnks also increased the levels of Myr-Axin-V5 in adult midguts ( S3B Fig ) . Furthermore , immunostaining revealed that even at these increased levels , Myr-Axin-V5 still localized predominately to the cell membrane of adult midgut enterocytes , suggesting that the subcellular distribution of Myr-Axin-V5 is not due to its lower levels ( S3C–S3E Fig ) . Together , these findings provided evidence that Tnks-dependent Axin proteolysis destabilizes Axin at the membrane . As Tnks is known to target Axin for proteolysis through ADP-ribosylation , we sought to determine if association with the membrane promotes Axin ADP-ribosylation , and thereby Axin degradation . To detect the low levels of ADP-ribosylated Axin , we took advantage of the finding that ADP-ribosylated Axin is recognized by the RING-type E3 ubiquitin ligase RNF146/Iduna for ubiquitination and subsequent degradation [13 , 14 , 44 , 45] . The WWE domain of RNF146 interacts directly with poly ( ADP-ribose ) in Axin to promote its ubiquitination . Pull downs using the WWE domain of RNF146 coupled to glutathione S-transferase ( GST ) specifically detect ADP-ribosylated Axin [14] . Therefore , to determine the level of ADP-ribosylated Axin in vivo , lysates from larvae expressing Axin-V5 were subjected to pull down with GST-WWE or the GST-WWER163A control , in which an arginine to alanine substitution abolishes interaction with poly ( ADP-ribose ) [14] . The pull down of Axin-V5 with GST-WWE , but not the GST-WWER163A control , confirmed the specificity of the assay ( Fig 5F ) . We further confirmed the specificity of the pull-down experiments through Tnks inactivation: ADP-ribosylated Axin-V5 was not detected in the GST-WWE pull-downs in lysates from Tnks null mutant larvae , in contrast with wild-type ( Fig 5G ) . To determine the extent to which the membrane-associated pool of Axin is ADP-ribosylated in vivo , we expressed Axin-V5 , Myr-Axin-V5 or MyrG-A-Axin-V5 in the wing discs of third instar larvae . As myristoylation of Axin-V5 results in markedly reduced levels , more larvae expressing Myr-Axin-V5 were used to obtain a comparable input for the GST-WWE pull-down assay . We found that the level of ADP-ribosylation of Myr-Axin-V5 was much higher than that of Axin-V5 or MyrG-A-Axin-V5 ( Fig 5H ) . These findings suggested that membrane association of Axin results in an increased pool of ADP-ribosylated Axin . As we had found that membrane association destabilizes Axin under basal conditions and that Tnks promotes the degradation of membrane-associated Axin ( Figs 5 and S3 ) , we sought to determine whether Tnks-mediated ADP-ribosylation is required for Axin membrane association . To test this hypothesis in vivo , we performed subcellular fractionation of lysates from wild-type and Tnks null mutant embryos within 2 . 5 hours of development ( prior to Wingless expression ) and determined the distribution of endogenous Axin in membrane and cytoplasmic fractions using immunoblots . We found that the subcellular distribution of Axin in Tnks null mutant embryos was similar to that in wild-type embryos , with pools of Axin in both the cytoplasmic and membrane fractions ( Fig 6A ) . This finding suggested that Tnks-mediated ADP-ribosylation is not essential for Axin membrane association . To test this conclusion using another approach , we investigated whether the Tnks binding domain of Axin is important for its membrane localization . We fractionated lysates from S2R+ cells expressing Axin-V5 or AxinΔTBD-V5 , and examined the distribution of Axin in immunoblots with V5 antibody . Equivalent levels of both Axin-V5 and AxinΔTBD-V5 were present in cytoplasmic and membrane compartments ( Fig 6B ) . These findings indicated that Tnks and the Tnks binding domain of Axin were dispensable for Axin membrane association . We conclude that ADP-ribosylation is not required for Axin membrane association , but that the membrane association of Axin results in increased ADP-ribosylation . Previous work based on Axin overexpression suggested that Wingless stimulation induced the bulk relocation of Axin from cytoplasm to cell membrane in Drosophila embryos [21 , 22] . Recently , we re-examined the effects of Wingless exposure in Drosophila embryos using an Axin-V5 transgene that is expressed within two-fold of endogenous levels , is able to replace the function of endogenous Axin , and does not disrupt Wnt signaling [18] . We discovered that Axin levels increase in Wingless-responding cells within thirty minutes of Wingless stimulation , resulting in a segmentally striped pattern of Axin in Drosophila embryos ( Fig 7A–7C ) , prior to Axin degradation hours later , and that this process is conserved in vertebrate cells [18] . We sought to use this in vivo system to examine the subcellular localization of Axin . Within the first three hours of embryogenesis , prior to the onset of Wingless expression , Axin was present both in the cytoplasm and at the plasma membrane , as revealed by its partial overlap with the transmembrane protein Neurotactin [46] ( S4 Fig ) . Notably , within thirty minutes following Wingless exposure , Axin puncta were observed near the cell periphery both in cells responding to Wingless stimulation and in those not exposed to Wingless ( Fig 7D and 7E ) . Importantly , we did not detect bulk redistribution of Axin from cytoplasm to membrane in Wingless responding cells , but instead found increased Axin signal in both the cytoplasm and at the plasma membrane ( Figs 7D , 7E and S5 ) . These results are in sharp contrast with a previous finding in which overexpressed Axin-GFP associated with the plasma membrane only in cells responding to Wingless stimulation [21] . We conclude that Wingless exposure induces an increase in Axin levels in both the cytoplasm and at the plasma membrane , and not bulk redistribution from cytoplasmic to membrane pools . Previous findings demonstrated that Axin levels increase in the membrane fraction following Wnt stimulation [3] . However , accompanying changes in the overall and cytosolic levels of Axin were not documented in that study , and thus it remained possible that the increase of Axin in the membrane fraction occurred concomitantly with an increase in overall Axin levels . To test this possibility , we used an approach that capitalized on our ability to detect endogenous Axin in lysates from cultured Drosophila embryonic cells . We treated S2R+ cells with Wingless conditioned medium ( Wg CM ) , and then subjected the lysates to subcellular fractionation and immunoblotting . Confirming robust Wingless pathway activation , the level of phosphorylated LRP6/Arrow increased rapidly following treatment with Wg CM ( Fig 8A , left panel ) . As reported previously , the total levels of Axin as determined by measuring all Axin isoforms increased rapidly though modestly following Wingless stimulation [18] ( Fig 8A , left panel , lane 1 and 4 ) . Furthermore , several slower migrating bands detected by the Axin antibody in the unstimulated state were not present after Wingless stimulation , suggesting an alteration in post-translational modification in response to pathway activation . This result is consistent with previous reports that mammalian Axin is dephosphorylated following Wingless exposure [5 , 24 , 25 , 47] . Importantly , supporting our immunostaining results in fly embryos , the relative ratio of Axin in the cytoplasmic and membrane fractions was similar in the absence or presence of Wingless stimulation , indicating that bulk redistribution of Axin from cytoplasm to membrane in response to Wingless does not occur ( Fig 8A , left panel ) . Our previous work revealed that after Wnt stimulation , the level of ADP-ribosylated Axin increases , and ADP-ribosylation promotes the interaction of Axin with phospho-LPR6/Arrow in both Drosophila and human cells [18] . To further investigate whether the cytosolic and membrane pools of Axin are differentially regulated in response to Wingless stimulation , we examined the subcellular localization of ADP-ribosylated Axin by subjecting cytosolic and membrane fractions from S2R+ cell lysates to the GST-WWE pull-down assay . Consistent with our hypothesis that Axin is ADP-ribosylated primarily at the membrane , we found that the vast majority of ADP-ribosylated Axin is present in the membrane fraction in the unstimulated state ( Fig 8A , right panel , lane 1 and 2 ) . We next determined whether Wingless exposure alters the subcellular localization of ADP-ribosylated Axin . We confirmed that Wingless stimulation induces a marked increase in the level of ADP-ribosylated Axin , and furthermore , found that this Wingless-induced increase in ADP-ribosylated Axin is confined largely to the membrane ( Fig 8A , right panel ) . Consistent with our previous finding that Axin ADP-ribosylation enhances its interaction with phospho-LRP6 [18] , phospho-LRP6 was also pulled down by GST-WWE in the membrane fraction following Wingless stimulation , likely through the interaction with ADP-ribosylated Axin ( Fig 8A , right panel ) . We conclude that Wingless exposure does not result in bulk redistribution of Axin from the cytosol to the plasma membrane , but rather induces increased levels of both cytoplasmic and membrane-associated Axin , with a preferential increase in the level of ADP-ribosylated Axin at the membrane , which likely facilitates the association between Axin and phospho-LRP6 and thus the activation of Wingless pathway .
Although Axin is a core component of Wnt pathway , the regulation of Axin remains poorly understood , largely due to technical challenges in studying endogenous Axin , which is present at low levels . Here , we capitalized on the sensitivity of an antibody that detects endogenous Drosophila Axin in vivo and in cultured cells to study Axin regulation . We found that even in the absence of Wnt stimulation , a pool of Axin is present in puncta that are localized at or proximal to the basolateral cell membrane in vivo . Subcellular fractionation of Drosophila embryos and cultured embryonic cells supported the membrane association of Axin in unstimulated conditions . These results were further supported by an experimental system in Drosophila embryos in which exogenous Axin was expressed at levels that are within two-fold of endogenous Axin and thus does not disrupt Wnt signaling [18] . We utilized this system to re-examine the proposed redistribution of Axin from cytosol to plasma membrane following Wingless stimulation . In contrast with previous work based on Axin overexpression at levels that abrogated Wingless signaling [21] , we found that the initial response to Wingless stimulation is not bulk translocation of Axin from cytosol to plasma membrane , but instead a modest increase in both cytoplasmic and membrane-associated Axin , and is followed by Axin proteolysis . Furthermore , evidence from our subcellular fractionation studies of Drosophila S2R+ cells responding to Wingless exposure also did not support the bulk relocation of Axin from cytoplasm to cell membrane in response to stimulation , but did suggest changes in Axin post-translational modification ( see below ) . Previous work indicated that Axin is sequestered in multivesicular bodies ( MVBs ) that form following Wnt stimulation [48]; however a subsequent study revealed that MVB formation is not required for Wnt pathway activation in Drosophila [49] . The membrane-proximal Axin puncta we have observed are unlikely to represent MVBs , since these puncta are found in the absence of Wnt exposure; however , the presence of these puncta may suggest that Axin associates with other types of vesicles juxtaposed with the plasma membrane . Previously , endogenous Axin was observed in puncta that were termed “degradasomes” in mammalian cultured cell lines; however these puncta were detected only when both Axin and Tnks levels were aberrantly increased by the use of Tnks inhibitors [50 , 51] . The presence of Axin in widespread puncta was also reported following overexpression of mammalian or Drosophila Axin and/or Apc , and those puncta were proposed to contain the destruction complex and promote β-catenin degradation [52–56] . Our documentation of membrane-proximal puncta containing Axin at endogenous levels furthers this previous work , and also reveals that the formation of these Axin puncta is largely dependent on the activity of endogenous Apc . As some of these Axin-containing puncta overlap the endogenous Apc signal , they may indeed represent the sites of destruction complex activity . How Apc promotes the localization of Axin to these membrane proximal puncta awaits further investigation , but may involve previously proposed roles for Apc in the facilitation of Axin multimerization [22 , 56] . We find that the membrane association of Axin promotes its proteolysis in the unstimulated state , and thus is important to maintain Axin at concentration-limiting levels . Previous studies revealed that Tnks-dependent Axin degradation is one of the mechanisms that maintain Axin at low levels in both mammalian and Drosophila cells [12 , 15 , 16] , and our results herein indicate that Tnks targets the membrane-associated pool of Axin for ADP-ribosylation and degradation . Therefore , we propose that Tnks-mediated ADP-ribosylation of Axin at the membrane is important to control Axin levels under basal conditions . Previous work revealed that mammalian Tnks localizes to the lateral membrane in polarized epithelial cells [57]; therefore , it is possible that enrichment of Tnks activity at the membrane promotes the proteolysis of membrane-associated Axin . Alternatively , membrane association may simply result in a local enrichment as Axin moves from a three-dimensional space ( the cytoplasm ) to a 2-dimensional surface ( the plasma membrane ) , and could thereby promote Tnks-dependent Axin ADP-ribosylation and subsequent degradation . Whether the cytoplasmic and membrane-associated pools of Axin are targeted for proteolysis through distinct mechanisms awaits further studies . Whereas ADP-ribosylated Axin is associated primarily with the membrane fraction , the membrane-association of Axin does not require ADP-ribosylation . In addition , three findings support the conclusion that the association of Axin with the membrane does not require Wingless stimulation: membrane-proximal Axin puncta are observed ubiquitously in larval imaginal discs , Axin is enriched at the cell membrane prior to the onset of Wingless expression during embryogenesis , and this membrane association is not disrupted by inactivation of the Wingless co-receptors LRP6/Arrow and Frizzled , both of which are essential for the response to Wingless stimulation . Previous studies in Xenopus embryos revealed that Axin at the plasma membrane could be partially precipitated using concanavalin A beads , indicating that a pool of Axin is associated with membrane glycoproteins [52] . The precise mechanisms that maintain Axin’s membrane association await further investigation . Taken together , our results indicate that a pool of Axin localizes to puncta at or near the cell membrane and is targeted for degradation by Tnks under basal conditions . We propose that this set up not only maintains Axin at low levels in the absence of Wnt stimulation , but also provides a mechanism that can trigger the rapid transition in Axin activity that facilitates signaling following Wnt stimulation . In support of this hypothesis , our recent work revealed that Axin levels , and in particular the ADP-ribosylated pool of Axin , increase rapidly after Wnt stimulation and that ADP-ribosylation enhances the association of Axin with phospho-LRP6/Arrow [18] . Herein , by analyzing the subcellular distribution of endogenous Axin , we found that the vast majority of ADP-ribosylated Axin is membrane-associated under basal conditions . In response to Wnt exposure , the increased pool of ADP-ribosylated Axin remains membrane-associated , and thus may further enhance the Axin-phospho-LRP6 interaction through both increased local concentration and increased affinity . The mechanisms underlying the rapid increase in Axin levels after Wnt exposure , and specifically in the ADP-ribosylated Axin pool , await further investigation , but we speculate that this response results from the rapid inhibition of Axin proteolysis that follows Wnt exposure . Furthermore , immunoblots with our Axin antibody suggest changes in post-translational modification of endogenous Axin in response to Wnt stimulation , consistent with the known Wnt-induced dephosphorylation of Axin that is thought to diminish the association between Axin and β-catenin [5 , 24 , 25] . Therefore , we hypothesize that modulation of two post-translational modifications in Axin—ADP-ribosylation as documented herein , and dephosphorylation as documented previously [5 , 24]—promote the initial response to Wnt stimulation . Specifically , dephosphorylation of Axin inhibits the destruction complex [24 , 47] , whereas ADP-ribosylation of the membrane-associated pool of Axin enhances the association of Axin with phospho-LRP6 , a step that promotes assembly of the signalosome [18] . Based on these findings , we propose a revised model in which the membrane association of Axin is important for its regulation in both unstimulated and stimulated states ( Fig 8B ) . In the absence of Wnt ligands , membrane-associated Axin is a substrate for ADP-ribosylation and ubiquitination , which targets Axin for degradation and thereby controls its limiting concentration that is important for the regulation of Wnt signaling . We speculate that Wnt stimulation rapidly inhibits Axin degradation , thereby inducing an increase in the level of membrane-associated , ADP-ribosylated Axin . As ADP-ribosylation enhances the interaction of Axin with phospho-LRP6 [18] , the Wnt-induced increase in levels of membrane-associated ADP-ribosylated Axin likely promotes the role of Axin in signaling activation . An extracellular gradient of Wingless protein forms on the basolateral surface of epithelial cells [58 , 59]; therefore we postulate that in response to Wingless stimulation , Axin associated with or near the basolateral cell membrane jump-starts the rapid association of Axin with LRP6 , which is among the earliest responses to Wnt exposure [3] .
A complete deletion of the Axin gene , Axin18 , was isolated by FLP-mediated trans-recombination between FRT sites [60] in PBac{RB}Mgat2e01270 and PBac{WH}Axnf01654 ( both obtained from the Exelixis collection at Harvard Medical School ) . Potential deletions were identified by lethal complementation tests with the mutant allele Axins044230 . Other stocks: Apc1Q8 [61] , Apc219 . 3 [62] , Ubx-FLP [63] , C765-Gal4 ( BDSC ) [64] , vestigial-Gal4 UAS-FLP [35] , FRT82B arm-lacZ [65] ( provided by J . Treisman , Skirball Institute , New York ) , cycAC8LR1 ubi-GFP FRT2A [66] , engrailed-Gal4 UAS-FLP ( provided by E . Piddini , NIMR , London ) , fz1P21 Dfz2C2 FRT2A [35] , FRT42D ubi-GFP PCNA775 [67] , hh-Gal4 UAS-FLP [66] , FRT42D arr2 [33] , Myo1A-Gal4 [68] , hsFLP1 [69] , 71B-Gal4 ( BDSC ) [64] , Axins044230 [28] , Tnks19 [17] , and Tnks503 [17] . The maternal α4-Gal4:VP16 driver ( mat-Gal4; line 67 ) contains the maternal tubulin promoter from αTub67C and the 3' UTR from αTub84B [70 , 71] . Canton S flies were used as wild-type controls . All crosses were performed at 25°C unless otherwise indicated . Somatic mitotic mutant clones were generated by FLP-mediated recombination [72] . Clones induced by hsFLP1 were generated by subjecting first and second instar larvae to a 37°C heat shock for 2hr and analyzed at third instar larval stage . Genotypes for generating mutant clones are as follows: Axin mutant clones: vestigial-Gal4 UAS-FLP/+; FRT82B Axin18/FRT82B arm-lacZ ( Fig 1 ) or hsFLP1/+; FRT82B Axin18/FRT82B arm-lacZ ( Fig 4 ) arrow mutant clones: FRT42D arr2/FRT42D ubi-GFP PCNA775; hh-Gal4 UAS-FLP/+ fz Dfz2 double mutant clones: en-Gal4 UAS-FLP/+; fzP21 Dfz2C2 FRT2A/cycAC8LR1 ubi-GFP FRT2A Apc1 or Apc2 mutant clones: Ubx-FLP/+; FRT82B Apc1Q8/FRT82B arm-lacZ or Ubx-FLP/+; FRT82B Apc219 . 3/FRT82B arm-lacZ S2R+ cells ( Drosophila Genomics Resource Center ) were maintained at 25°C in Schneider’s complete medium: Schneider’s Drosophila medium with L-glutamine ( Gibco ) supplemented with 10% FBS ( Gibco ) and 0 . 1mg/mL penicillin/streptomycin ( Invitrogen ) . Cells were transiently transfected using calcium-phosphate DNA precipitation [73] . To collect Wingless conditioned medium ( Wg CM ) , S2TubWg cells ( Drosophila Genomics Resource Center ) were grown to confluence , then split 1:3 and incubated at 25°C for 72 hours . Cells were then re-suspended in the medium and centrifuged at 1000 x rpm for 5 minutes at room temperature; the supernatant was centrifuged again at 5000 x rpm for 5 minutes at room temperature . The resulting supernatant contained the Wingless conditioned medium , which was stored at 4°C . For treatment with Wg CM , cells were washed one time with serum-free , antibiotic-free , Schneider’s medium . Wg CM or complete medium ( CTR ) was added and cells were incubated at 25°C for 1 hour . pUASTattB-Axin-V5 and pUASTattB-AxinΔTBD-V5 were generated as described [18] . To generate pUASTattB-Myr-Axin-V5 transgene , the myristoylation sequence was added to pUASTattB-Axin-V5 [18] by two rounds of PCR-mediated mutagenesis using the oligonucleotides: forward: 5’-ATG GGC AAC AAA TGC TGC AGC AAG CGA CAG AGT TTC atg agt ggc cat cca tcg gga at-3’ ( residues in upper-case denote the myristoylation sequence ) , and reverse: 5’-gtc aac ttc ctc gag cag-3’ . The resulting PCR product was used as a template and amplified with the oligonucleotides: forward: 5’-gag ggt acc tac tag tcc agt gtg gtg gaa ttg atc ATG GGC AAC AAA TGC TGC AGC AA -3’ , and reverse: 5’-gtc aac ttc ctc gag cag-3’ . The resulting fragment was digested with KpnI and XhoI and inserted into pUASTattB-Axin-V5 at the KpnI and XhoI sites . The control pUASTattB-MyrG-A-Axin-V5 was generated by PCR-mediated mutagenesis using pUASTattB-Myr-Axin-V5 as a template and the oligonucleotides: forward: 5’- gag ggt acc tac tag tcc agt gtg gtg gaa ttg atc ATG GCC AAC AAA TGC TGC AGC AA -3’ and reverse: 5’-gtc aac ttc ctc gag cag-3’ . The resulting fragment was digested with KpnI and XhoI and inserted into pUASTattB-Axin-V5 at the KpnI and XhoI sites . To generate pUASTattB-Myr-AxinΔTBD-V5 , the same procedure and the same oligonucleotides as above were used . The resulting fragment was digested with KpnI and XhoI and inserted into pUASTattB-AxinΔTBD-V5 at the KpnI and XhoI sites . Plasmids used for transfection of Drosophila S2R+ cells were pAc5 . 1-Axin-V5 , pAc5 . 1-AxinΔTBD-V5 , pAc5 . 1A-Myr-Axin-V5 and pAc5 . 1-Myr-AxinΔTBD-V5 . To generate the plasmids: fragments encoding Axin-V5 , AxinΔTBD-V5 , Myr-Axin-V5 , Myr-AxinΔTBD-V5 from pUASTattB-Axin-V5 , pUASTattB-AxinΔTBD-V5 , pUASTattB-Myr-Axin-V5 and pUASTattB-Myr-AxinΔTBD-V5 respectively , were digested using KpnI and XbaI . The resulting fragments were inserted into the pAc5 . 1 A vector ( Invitrogen ) at the KpnI and XbaI sites . For S2R+ cell lysates used in immunoblots , cells were washed with cold PBS and lysed in 4X Laemmli buffer supplemented with 0 . 1M DTT . For larval lysates used in immunoblots , third instar larvae were dissected to remove salivary glands , fat body , and gut tissues in cold PBS . After removal of PBS , 4X Laemmli loading buffer supplemented with 0 . 1M DTT was added and the lysates were vortexed briefly . For midgut lysates used in immunoblots , midguts from 5-day-old adults were dissected in cold PBS , and then treated with 1x Trypsin EDTA ( Corning Life Sciences ) for 2 hours at room temperature . Tissues were washed with PBS and homogenized in 4X Laemmli loading buffer supplemented with 0 . 1M DTT . All the lysates were incubated for 5 minutes at 100°C before SDS-PAGE analysis . Quantification of immunoblots was performed with ImageJ ( Wayne Rasband , National Institutes of Health ) . Statistical analysis ( t-test ) was performed using Prism ( GraphPad ) . For immunostaining , third instar larval wing imaginal discs and pupal wings were dissected in PBS , fixed in 4% paraformaldehyde in PBS for 20 minutes . For immunostaining of adult guts , midguts were dissected in PBS , fixed in 4% paraformaldehyde in PBS for 45 minutes . After fixation , all samples were washed with PBS with 0 . 1% Triton X-100 , followed by incubation in PBS with 0 . 5% Triton X-100 and 10% BSA for 1 hour at room temperature . Incubation with primary antibodies was performed at 4°C overnight in PBS with 0 . 5% Triton X-100 . Incubation with secondary antibodies was for 2 hours at room temperature . Specimens were mounted in Prolong Gold ( Invitrogen ) . Immunostaining of embryos was performed as described [18] . Fluorescent images were obtained on a Nikon A1RSi confocal microscope or Zeiss LSM 880 microscope with Airyscan ( Fig 2G–2I ) and processed using Adobe Photoshop software . The primary antibodies used for immunoblotting were mouse anti-V5 ( 1:5000 , Invitrogen ) , guinea pig anti-Axin ( 1:1000 , [17] ) , rabbit anti-Kinesin Heavy Chain ( 1:10000 , Cytoskeleton ) , mouse anti-alpha-Tubulin ( 1:10000 , DM1A , Sigma ) , rabbit anti-alpha-Tubulin ( 1:10000 , Sigma ) , rabbit anti-Gluthathione-S-Transferase ( 1:10000 , Invitrogen ) , rabbit anti-phospho-LRP6 [Thr1572] ( 1:1000 , Millipore ) , mouse anti-Nervana antibody ( Nrv5F7 , 1:1000 , DSHB ) , and guinea pig anti-Arrow antibody ( 1:1000 , [74] ) . The primary antibodies used for immunostaining were guinea pig anti-Axin ( 1:1000 , [17] ) , rabbit anti-β-gal ( 1:1000; MP Biomedicals ) , mouse anti-Arm ( 1:20; DSHB ) , mouse anti-Fas III ( 1:20; DSHB ) , mouse anti-Discs Large ( 1:20; DSHB ) , rabbit anti-GFP ( 1:200; Invitrogen ) , mouse anti-V5 ( 1:5000; Invitrogen ) , rabbit anti-Apc2 ( 1:1000; [75] ) , and guinea pig anti-Senseless ( 1:1000 , [76] ) . The secondary antibodies used for immunoblotting were: goat anti-rabbit HRP conjugate ( 1:10000 , Biorad ) , goat anti-mouse HRP conjugate ( 1:10000 , Biorad ) , and goat anti-guinea pig HRP conjugate ( 1:10000 , Jackson ImmunoResearch ) . The secondary antibodies used for immunostaining were goat or donkey Alexa Fluor 488 , 555 or Cy5 conjugates ( 1:400; Invitrogen ) . Embryos were collected 0–2 . 5 hours after egg lay , dechorionated in bleach for 45 seconds , and washed extensively with water and 1X PBS . Embryos and S2R+ cells were lysed in lysis buffer ( 20mM HEPES , 10mM KCL , pH 7 . 9 ) supplemented with 0 . 5mM DTT and 1X protease/phosphatase inhibitor cocktail ( Pierce ) with a dounce homogenizer ( 200 strokes ) . Lysates were spun at 1000 x g for 10 minutes to obtain total lysate . Supernatant was subsequently spun at 100 , 000 x g for 30 minutes to pellet the membrane fraction . Supernatant containing cytosolic fraction was saved and pellet containing membrane fraction was resuspended in lysis buffer supplemented with 0 . 5mM DTT , 1% NP-40 , and 1X protease/phosphatase inhibitor cocktail ( Pierce ) . For GST pull downs , GST-WWE and GST-WWER164A beads were generated as described previously [14] . S2R+ cells were treated as indicated , then washed once with cold 1X PBS and lysed in RIPA buffer ( 50mM Tris [pH 8 . 5] , 300 mM NaCl , 1% NP-40 , 0 . 5% sodium deoxycholate , and 0 . 1% SDS ) supplemented with 1μM ADP-HPD ( Enzo Lifesciences ) , and 1X protease and phosphatase inhibitor cocktail ( Pierce ) . Lysates were incubated with GST-WWE or GST-WWER164A beads overnight at 4°C . Following incubation , beads were washed four times in wash buffer ( 50mM TrisHCl [pH 8 . 0] , 150mM NaCl , 1% NP-40 , 10% Glycerol , 1 . 5mM EDTA [pH 8 . 0] ) supplemented with 1μM ADP-HPD and 1X protease and phosphatase inhibitor cocktail ( Pierce ) . Bound materials were eluted with 4X sample buffer and resolved by SDS-PAGE , transferred to nitrocellulose membranes and blotted with the indicated antibodies . | Axin is a scaffold protein with essential roles in Wnt signal transduction . In the classical model , the transition from the unstimulated to stimulated state is thought to be achieved by recruitment of Axin from cytosol to plasma membrane . We find that a pool of endogenous Drosophila Axin is localized in puncta juxtaposed with the cell membrane even under basal conditions and is targeted for degradation by the ADP-ribose polymerase Tankyrase . Wnt stimulation initially results in increased Axin levels in both the cytosolic and membrane pools , which may enhance the robust activation of signaling . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"invertebrates",
"molecular",
"probe",
"techniques",
"membrane",
"staining",
"immunoblotting",
"cloning",
"animals",
"animal",
"models",
"developmental",
"biology",
"drosophila",
"melanogaster",
"model",
"organisms",
"molecular",
"biology",
"techniques",
"embryos",
"cellular",
"structures",
"and",
"organelles",
"adp-ribosylation",
"drosophila",
"research",
"and",
"analysis",
"methods",
"specimen",
"preparation",
"and",
"treatment",
"staining",
"embryology",
"chemistry",
"molecular",
"biology",
"insects",
"cell",
"membranes",
"arthropoda",
"cytoplasm",
"cell",
"biology",
"biology",
"and",
"life",
"sciences",
"chemical",
"reactions",
"physical",
"sciences",
"organisms"
] | 2016 | Dual Roles for Membrane Association of Drosophila Axin in Wnt Signaling |
Ribose 5-phosphate isomerase is an enzyme involved in the non-oxidative branch of the pentose phosphate pathway , and catalyzes the inter-conversion of D-ribose 5-phosphate and D-ribulose 5-phosphate . Trypanosomatids , including the agent of African sleeping sickness namely Trypanosoma brucei , have a type B ribose-5-phosphate isomerase . This enzyme is absent from humans , which have a structurally unrelated ribose 5-phosphate isomerase type A , and therefore has been proposed as an attractive drug target waiting further characterization . In this study , Trypanosoma brucei ribose 5-phosphate isomerase B showed in vitro isomerase activity . RNAi against this enzyme reduced parasites' in vitro growth , and more importantly , bloodstream forms infectivity . Mice infected with induced RNAi clones exhibited lower parasitaemia and a prolonged survival compared to control mice . Phenotypic reversion was achieved by complementing induced RNAi clones with an ectopic copy of Trypanosoma cruzi gene . Our results present the first functional characterization of Trypanosoma brucei ribose 5-phosphate isomerase B , and show the relevance of an enzyme belonging to the non-oxidative branch of the pentose phosphate pathway in the context of Trypanosoma brucei infection .
African sleeping sickness is a vector borne disease of mammals , caused by Trypanosoma brucei ( T . brucei ) , for which the development of more effective , safe , and affordable chemotherapies remains a major goal . Vaccines are unlikely to be suitable [1]–[3] , and therefore disease control relies exclusively on chemotherapy . The glucose-based metabolism is a key metabolic pathway for bloodstream forms , the mammalian infective stages . The absence of a fully functional mitochondrion along with a remarkable high proliferation rate makes parasites entirely dependent on glucose [4] , [5] . The glucose-based metabolism comprises two pathways: the glycolytic pathway and the pentose phosphate pathway ( PPP ) . Despite using the same substrate , the pathways have different functions . Glycolysis catabolizes glucose for ATP requirements , while PPP includes an oxidative branch , mainly involved in the maintenance of cell redox homeostasis , and a non-oxidative branch in which ribose 5-phosphate is produced for nucleotide and nucleic acid synthesis . Enzymes involved in the PPP non-oxidative branch include ribose-5-phosphate isomerase , ribulose-5-phosphate epimerase , transaldolase and transketolase , and in contrast with enzymes involved in the glycolysis [6]–[15] or in the oxidative PPP [16] , [17] , have been less studied . In T . brucei , enzymes of the non-oxidative branch downstream ribose-5-phosphate isomerase are apparently developmentally regulated [18] . Ribose 5-phosphate epimerase and transketolase activities were only detected in procyclics , the parasite form present in the insect vector . This suggests that in the mammalian host , bloodstream forms constrain sugar metabolism to the production of ribose-5-phosphate and NADPH via the oxidative phase of the PPP , most likely to meet the remarkably high proliferation rate of these parasites [19] , and/or to protect themselves against a variety of reactive oxygen and nitrogen species [20] , [21] in a context of an in vivo infection . Ribose-5-phosphate isomerase ( Rpi ) catalyzes the inter-conversion between ribulose-5-phosphate ( Ru5P ) and ribose 5-phosphate ( R5P ) . Contrary to trypanosomatids , which have a Rpi type B ( RpiB ) , the presence of a structurally unrelated Rpi type A ( RpiA ) in humans together with the adverse phenotype observed in rpiA-/rpiB- knockout Escherichia coli ( E . coli ) [22] have led to suggest RpiB as an attractive drug target candidate that waits further characterization . In this study , we investigate the importance of RpiB in T . brucei bloodstream form viability and infectivity .
All experiments were carried out in accordance with the IBMC . INEB Animal Ethics Committees and the Portuguese National Authorities for Animal Health guidelines , according to the statements on the directive 2010/63/EU of the European Parliament and of the Council . IL , JT and ACS have an accreditation for animal research given from Portuguese Veterinary Direction ( Ministerial Directive 1005/92 ) . Procyclic and bloodstream T . brucei Lister 427 were cultivated in MEM-Pros and HMI-9 medium , respectively , as previously described [23] . Bloodstream forms containing pHD1313 [24] were maintained with 0 . 2 µg/ml phleomycin . Ribose 5-phosphate isomerase B genes from T . brucei ( TbRPIB ) and T . cruzi ( TcRPIB ) were obtained by performing PCR on genomic DNA from Trypanosama brucei TREU927 and Trypanosoma cruzi CL Brener Non-Esmeraldo-like . Fragments of the open reading frames of TbRPIB ( Tb927 . 11 . 8970; chromosome Tb927_11_v5 from 2 , 462 , 183 to 2 , 463 , 307 ) and TcRPIB ( Tc00 . 1047053508601 . 119; chromosome TcChr30-P from 475 , 724 to 476 , 203 ) were PCR-amplified using a Taq DNA polymerase with proofreading activity ( Roche ) . The primers were as follows: sense primer 5′ - CAATTTCCATATGACGCGCAAGGTGGC - 3′ and antisense primer 5′ - CCCAAGCAAGCTTCTAACAACCATTCG - 3′ , sense primer 5′ - CAATTTCCATATGACGCGCCGAGTCGC - 3′ and antisense primer 5′ - CCCAAGCGAATTCTCATTTTACCCCTTTG - 3′ , respectively . PCR conditions were as follows: initial denaturation ( 2 min at 94°C ) , 35 cycles of denaturation ( 30 s at 94°C ) , annealing ( 30 s at 40°C ) and elongation ( 2 min at 68°C ) followed by a final extension step ( 10 min at 68°C ) ; initial denaturation ( 2 min at 94°C ) , 35 cycles of denaturation ( 30 s at 94°C ) , annealing ( 30 s at 58°C ) elongation ( 2 min at 68°C ) and a final extension step ( 10 min at 68°C ) , respectively . The PCR products were isolated from a 1% agarose gel , purified by the Qiaex II protocol ( Qiagen ) , and cloned into a pGEM-T Easy vector ( Promega ) and sent to Eurofins MWG ( Germany ) for sequencing . All fragments were checked against the T . brucei and T . cruzi genome sequence database ( http://www . genedb . org ) using Blast to ensure their specificity . The TbRPIB and TcRPIB genes were excised from the pGEM-T Easy vector ( using NdeI/EcoRI restriction enzyme combination ) , gel purified and subcloned into pET28a ( + ) expression vector ( Novagen ) . The resulting constructs presented a poly-His tag ( 6× Histidine residues ) at the N-terminal and were used to transform E . coli BL21DE3 cells . Both recombinant proteins were expressed by induction of log-phase cultures ( 500 ml; OD600 = 0 . 6 ) with 0 . 5 mM IPTG ( isopropyl-β-D- thiogalactopyranoside ) for 3 h at 37°C and agitation at 250 rpm/min . Bacteria were harvested by centrifugation ( 4000 rpm , for 40 min , at 4°C ) , resuspended in 20 ml of buffer A ( 0 . 5 M NaCl , 20 mM Tris . HCl , pH 7 . 6 ) . The sample was sonicated , according to the following conditions: output 4 , duty cycle 50% , 10 cycles with 15 s each . Centrifugation ( 4000 rpm , for 60 min , at 4°C ) was followed to obtain the bacterial crude extract . The recombinant enzymes were purified in one step using Ni2+ resin ( ProBond ) pre-equilibrated in buffer A . The column was washed sequentially with 2–3 ml of the buffer A , 20 ml of the bacterial crude extract , 2 ml of buffer A 25 mM imidazole , 2 ml of buffer A 30 mM imidazole , 2 ml of buffer A 40 mM imidazole , 2 ml of buffer A 40 mM imidazole , 2 ml of buffer A 50 mM imidazole , 10 ml of buffer A 100 mM imidazole , 5 ml of buffer A 500 mM imidazole and 8 ml of buffer B ( 1 M imidazole , 0 . 5 M NaCl , 200 mM Tris , pH 7 . 6 ) . TbRpiB and TcRpiB were eluted in the fractions of buffer A containing between 100 and 500 mM of imidazole . Dialysis was performed against 100 mM Tris/HCl ( pH 7 . 6 ) . To generate rat polyclonal antibody against TbRpiB , and rabbit polyclonal antibodies against TbRpiB and TcRpiB , each animal was first immunized with 150 µg of recombinant protein . After 2 weeks , 4 boosts with 100 µg of recombinant TbRpiB or TcRpiB were given weekly . The collected blood samples were centrifuged to obtain the sera . Multiple sequence alignments were performed in ClustalW [25] and images prepared with Aline , Version 011208 [26] . Homology models were obtained in SWISS-MODEL , using PDB accession code 3K7S as a template [27]–[29] . 3D structures were rendered with PyMOL ( The PyMOL Molecular Graphics System , Version 1 . 3 , Schrödinger , LLC ) . TbRpiB activity was assessed through Km determination for R5P and Ru5P , through 4-deoxy-4-phospho-D-erythronohydroxamic acid ( 4-PEH ) ( kindly provided by Dr . Laurent Salmon ) inhibitory capacity against TbRpiB , and through 4-PEH inhibition mechanism characterization . Firstly , to determine the Km for R5P and to characterize 4-PEH-inhibition mechanism , a direct spectrophotometric method at 290 nm [30] was used , to quantify Ru5P formation . Km determination was performed at R5P concentrations in a range between 3 . 1 and 50 mM in Tris/HCl ( pH 7 . 6 ) . For 4-PEH inhibition mechanism characterization , the experiment was performed in the presence of 0 . 5 µg of enzyme and 0 . 1 , 0 . 4 , 0 . 7 or 1 mM of inhibitor . All inhibitors were tested in the presence of 3 . 1 mM R5P . A negative control was made using heat inactivated enzyme . The TcRpiB enzyme was used as a positive control [31] . A calibration curve for Ru5P , using the referred method , was established to determine enzyme activity . An absorbance of 0 . 0381 at 290 nm was considered for 1 mM Ru5P . To determine the Km for Ru5P and to test 4-PEH inhibition as well , a modification of Dische's Cysteine-Carbazole method was used [32] . To determine Km , an incubation mixture contained 5 µl of 0 . 05 µg of enzyme in buffer A [100 mM Tris/HCl ( pH 8 . 4 ) , 1 mM EDTA and 0 . 5 mM 2-mercaptoethanol] plus 5 µl of Ru5P , giving final concentrations between 0 . 625 and 10 mM Ru5P , was used . For inhibition assay , Ru5P concentration used was 1 . 25 mM . Incubation was done for 10 min at room temperature . Following incubation , 15 µl of 0 . 5% cysteinium chloride , 125 µl of 75% ( v/v ) sulfuric acid and 5 µl of a 0 . 1% solution of carbazole in ethanol were added . After 30 min standing at room temperature , the A546 was determined . A blank without enzyme was run for each substrate or inhibitor concentration . Reaction linearity was checked varying enzyme concentration and time . To estimate the remaining Ru5P , a calibration curve was generated . In this assay conditions , 1 mM of Ru5P gave an A546 of 0 . 270 in a final reaction volume of 155 µl . For anti-TbRpiB antibodies validation , cells from log-phase cultures of T . brucei RNAi cell lines and wt strain were centrifuged and resuspended at 106/ml in PBS . The cells were fixed in µ-Chamber 12 well ( Ibidi ) for 15 min , at room temperature , in PBS containing 4% p-formaldehyde , washed twice with PBS , and then permeabilized in PBS containing 0 . 1% of Triton X-100 . The coverslips were incubated in PBS containing 10% FCS during 60 min , at room temperature , in a humidified atmosphere and washed twice with PBS/2% FCS . Then , incubated with primary rat or rabbit polyclonal antibodies against TbRpiB ( 1∶100 and 1∶1000 respectively , both diluted in blocking solution ) overnight , at 4°C , followed by two washes with PBS/2% FCS ( 5 min each one ) . Subsequently , cells were incubated with Alexa Fluor 647 conjugated goat anti-rat or Alexa Fluor 488 conjugated goat anti-rabbit secondary antibodies ( Molecular probes from Life technologies ) ( 1∶500 diluted in blocking solution ) for 1 h at room temperature in an humidified atmosphere , then washed twice with PBS . The coverslips were then stained and mounted with Vectashield-DAPI ( Vector Laboratories , Inc . ) . Images were captured using fluorescence microscope AxioImager Z1 and software Axiovision 4 . 7 ( Carl Zeiss , Germany ) . Pseudo-coloring of images were carried out using ImageJ software ( version 1 . 43u ) . In case of TbRpiB immunolocalization , bloodstream form T . brucei wt cells were probed using primary rat anti-TbRpiB ( 1∶100 diluted in blocking solution ) and primary rabbit polyclonal antibody against aldolase ( glycosome marker , 1∶5000 diluted in blocking solution ) . Cells were then incubated with biotin conjugated goat anti-rat ( 1∶500 diluted in blocking solution ) ( BD Pharmingen ) for 1 h room temperature in a humidified atmosphere , then washed twice with PBS/2% FCS . Subsequently , cells were incubated with Alexa Fluor 647 conjugated goat anti-rabbit ( Molecular probes , Life technologies ) and Streptavidin-FITC ( BD Pharmingen ) secondary antibodies ( 1∶1000 diluted in blocking solution ) for 1 h at room temperature in an humidified atmosphere , then washed twice with PBS . Vertical stacks were captured , using an confocal microscope Leica TCS SP5II and LAS 2 . 6 software ( Leica Microsystems , Germany ) . Mean fluorescence intensity of aldolase and RpiB was determined in each stack for the projected co-localization areas . Quantifications were carried out using ImageJ software ( version 1 . 43u ) . For each sample condition , bloodstream cells were washed once with cold trypanosome homogenisation buffer ( THB ) , composed by 25 mM Tris , 1 mM EDTA and 10% sucrose , pH = 7 . 8 . Just before cell lyses , leupeptin ( final concentration of 2 µg/ml ) and different digitonin quantities ( final concentrations of 5 , 12 . 5 , 25 , 50 , 100 , 150 and 200 ug/ml ) were added to 500 µl of cold THB , for cell pellet resuspension . Untreated cells ( 0 µg/ml of digitonin ) and those completely permeabilized ( total release , the result of incubation in 0 . 5% Triton X-100 ) were used for comparison . Each sample condition was incubated 60 min on ice , and then centrifuged at 2000 rpm , 4°C , for 10 min . Supernatants were taken and 500 µl of cold THB was added to each pellet . All fractions were analysed through Western blot for Rpi ( 108 cells per well; 1∶1000 polyclonal rabbit anti-TbRpiB as primary antibody ) , enolase ( 107 cells per well; 1∶5000 polyclonal rabbit anti-enolase as primary antibody ) and aldolase ( 107 cells per well; 1∶5000 polyclonal rabbit anti-aldolase as primary antibody ) . HRP-conjugated goat anti-rabbit ( 1∶5000 ) was used as secondary antibody . TbRPIB fragment ( sense oligo with a BglII – SphI linker 5′ – GAGAAGATCTGCATGCGCGCAAGGTGGCTATCGGTG - 3′ , and an antisense oligo with a ClaI – SalI 5′ – GCTAGCTACAGCTGACGGTCCTCCCCGCTGTATG – 3′ ) was cloned twice in opposite direction on either sides of a “stuffer” of the pHD1144 vector . The resulting construct obtained through HindIII and BglII digestion was cloned into pHD1145 . The final construct was transfect into bloodstream forms with pHD1313 , and stable individual clones were selected with 7 . 5 µg/ml of hygromycin . For functional complementation , TcRPIB fragment ( sense oligo with a HindIII linker 5′ - GAAGCTTATGACGCGCCGAGTCGCAAT - 3′ , and an antisense oligo with a BglII linker 5′ - AGATCTTCATTTTACCCCTTTGTTCC - 3′ ) , was cloned in pHD1034 vector ( digested with HindIII and BamHI ) . After transfection [33] , individual clones were selected with 0 . 2 µg/ml of puromycin . For in vitro growth curves , cell lines were seeded at 2×105 parasites/ml of complete HMI-9 medium , in the absence and presence of 100 ng/ml of tetracycline ( tet ) . Every 24 h , until day 10 , cell growth was monitored microscopically . For in vivo infections , after 24 h in the absence of selective drugs , and then a further 48 h of tet induction , 104 wt and transgenic parasites were inoculated intraperitoneally in 6–8 weeks old BALB/c mice ( n = 3–8 ) . 48 h prior infection , the RNAi induced mice were treated with 1 mg/ml doxycycline hyclate and 5% sucrose containing water [34] , while RNAi non-induced mice were given standard water . Parasitaemia was measured daily from the six day post-infection through tail blood extraction , during a period which all mice in the group were alive . Total RNA was isolated from ≈2×107 bloodstream forms using Trizol reagent ( Life Technologies ) . 10 µg RNA were directly separated by overnight formaldehyde agarose-gel electrophoresis , transferred onto a nylon membrane by capillarity and fixed by UV irradiation . The membrane was prehybridized in a hybridization bottle in 5× SSC , 0 . 5% SDS with salmon sperm DNA ( 200 µg/ml ) and 1× Denhardt's solution for 2 hours at 65°C . TbRPIB and signal recognition particle ( SRP; Tb927 . 8 . 2861_7SL ) probes were generated by PCR in the presence of [32P]-labelled dCTP using Prime-It RmT random primer labelling kit ( Stratagene ) followed by purification using QIAquick Nucleotide Removal Kit ( QIAGEN ) . Denaturated radioactive probes were added to the prehybridization solution at 65°C and incubated overnight . After rinsing the membrane twice for 5 min . with 2× SSC/0 . 1% SDS , the probes were washed out with two washes of 30 minutes in 0 . 1× SSC/0 . 1% SDS at 65°C and the membrane exposed on a Fugifilm FLA-3000 reader screen . ImageJ software ( version 1 . 43u ) was used for RNA quantification . Cell free extracts were obtained in RIPA buffer ( 20 mM Tris-HCl ( pH 7 . 5 ) , 150 mM NaCl , 1 mM Na2EDTA , 1 mM EGTA , 1% NP-40 , 1% sodium deoxycholate , 2 . 5 mM sodium pyrophosphate , 1 mM b-glycerophosphate , 1 mM Na3VO4 ) , with freshly-added complete protease inhibitor cocktail ( Roche Applied Science ) . The total protein amount was quantified using Biorad Commercial Kit ( Reagents A , B and S ) and the samples were then kept at -80°C . For analysis of parasites collected from mice , trypanosomes were purified from mouse blood using a DE-52 ( Whatman ) column [35] . For Western blotting , 10 µg of recombinant TbRpiB and TcRpiB proteins were resolved in 15% SDS/PAGE ( Tris-Tricine gel ) , while 30 µg of total soluble cell extract and 107 parasites were resolved in 12% Tris-Glycine SDS/PAGE , and all were then transferred on to a nitrocellulose Hy-bond ECL membrane ( Amersham Biosciences ) . The membrane was blocked in 5% ( w/v ) non-fat dried skimmed milk in PBS/0 . 1% Tween-20 ( blocking solution ) , followed by incubation with an anti-His-tag rabbit antibody ( MicroMol-413 ) ( 1∶1000 ) or a combination of an anti-TbRpiB rabbit antibody ( 1∶1000 ) with an anti-aldolase rabbit antibody ( 1∶5000 ) in blocking solution at 4°C overnight , respectively . Blots were washed with PBS/0 . 1% Tween-20 ( 3 times 15 min ) . Horseradish peroxidase-conjugated goat anti-rabbit IgG ( Amersham ) ( 1∶5000 for 1 h , at room temperature ) was used as the secondary antibody . The membranes were developed using SuperSignal WestPico Chemiluminescent Substrate ( Pierce ) . ImageJ software ( version 1 . 43u ) was used for protein bands semi-quantification . Student's t-test and Graphpad Prism Software ( version 5 . 0 ) were used . p values ≤0 . 05 were considered to be statistically significant ( * p≤0 . 05 , ** p≤0 . 01 , *** p≤0 . 001 ) .
An open reading frame with sequence similarity to RpiB was identified both in T . brucei ( Tb927 . 11 . 8970 ) and in T . cruzi ( Tc00 . 1047053508601 . 119 ) genomes . Protein sequence alignment using ClustalW [25] revealed 67% identity for TbRpiB versus TcRpiB , and both proteins show no similarity with human ribose 5–phosphate isomerase A . TcRpiB and TbRpiB contain 159 and 155 amino acids residues per monomer , respectively . Protein multiple sequence alignment of RpiB from T . cruzi CL Brener Esmeraldo-like ( Tc00 . 1047053509199 . 24; PDB accession code 3K7S [36] ) , T . cruzi CL Brener Non-Esmeraldo-like ( Tc00 . 1047053508601 . 119 ) and T . brucei ( Tb927 . 11 . 8970 ) is shown in S1A Fig . The scale colour , from cyan ( low-similarity residues ) to red ( high-similarity residues ) , underlines the degree of similarity between the three protein sequences , also seen in the TcRpiB ( Esmeraldo like strain ) ribbon representation ( S1B Fig . ) . The superposition of TcRpiB ( Esmeraldo like strain ) structure ( grey ) ( PDB code 3K7S ) , with the homology models generated for TcRpiB ( Non Esmeraldo like strain ) ( purple ) and TbRpiB ( blue ) show a high structural homology and strict conservation of the residues involved in R5P binding pocket ( S1A , C Fig . ) . Biochemical studies were performed using histidine-tagged fusion TbRpiB and TcRpiB ( positive control ) proteins expressed in E . coli and purified under non-denaturing conditions ( Figs . 1A , S2A ) . The T . brucei and T . cruzi [31] enzymes have in vitro ribose 5-phosphate isomerase activity , as these proteins can use both R5P and Ru5P as substrates . For R5P , T . brucei protein showed a significantly higher Km ( 2 . 8 fold increase , p<0 . 05 ) , but not a lower maximum velocity ( Vmax ) or catalytic constant ( kcat ) compared to T . cruzi enzyme ( Table 1 and S2B Fig . ) . For Ru5P , the Km of the T . brucei protein was not significantly different from that of the T . cruzi enzyme value , but the Vmax and kcat were higher ( ≈1 . 5 fold , p<0 . 05 ) ( Table 1 and S2B Fig . ) . Both the T . brucei and the T . cruzi enzymes exhibited significant lower Kms for Ru5P than for R5P , ( 5 . 2 fold , p<0 . 05 and 3 . 7 fold , p<0 . 01 , respectively ) , suggesting the reaction occurs preferentially from Ru5P to R5P . The turnover values ( kcat ) were found to be significantly higher for Ru5P than for R5P , in both T . brucei ( p = 0 . 001 ) and T . cruzi ( p<0 . 001 ) enzymes ( Table 1 and S2B Fig . ) . The reaction mechanism of ribose 5-phosphate isomerase involves two steps: an initial opening of the furanose ring of R5P , followed by the aldolase-ketose isomerisation , via a cis-enediolate high energy intermediate [31] . 4-PEH has been described to act as a competitive inhibitor which compromises the binding of 1 , 2-cis-enediolate intermediate [37] . The inhibitory capability of 4-PEH was screened in vitro , resulting in an IC50 of 0 . 8 mM and 0 . 7 mM for TbRpiB ( Fig . 1B ) and TcRpiB ( S2C Fig . ) , respectively , with Ki values of 2 . 2 ( Fig . 1C ) and 1 . 6 mM ( S2D Fig . ) . 4-PEH showed , as expected , a competitive inhibition behaviour , once using increasing concentrations of inhibitor , a progressive increase in the Km for R5P without Vmax alteration was observed ( Figs . 1D , S2E ) . The inhibitor behaviour , and also the IC50 and the Ki values are in agreement to what was described before for T . cruzi enzyme [31] , [36] . 4-PEH was also reported as a potent inhibitor against Mycobacterium tuberculosis RpiB [37] . Undoubtedly , TbRpiB has isomerase activity and uses preferentially ribulose 5-phosphate as a substrate . Rabbit and rat polyclonal antibodies were generated against the TbRpiB recombinant protein . Antibody specificity was validated , as induction of RpiB RNAi resulted in a decrease in the fluorescence intensity of bloodstreams when compared to non-induced parasites ( S3A , B , C Fig . ) . Similarly a significant decrease on RpiB levels in the extracts of TbRpiB RNAi induced parasites is shown by Western blot . Rat and rabbit antibodies specificity against RpiB can be appreciated on the whole Western blot membranes ( S3D , E Fig . ) . Using rabbit polyclonal antibody against parasite extracts , TbRpiB was found more abundant in procyclic forms than in bloodstream forms ( Fig . 2A ) . To ascertain RpiB subcellular localization in bloodstream forms , two complementary approaches , immunofluorescence and digitonin fractionation , were performed . Fluorescent confocal microscopy analysis suggests that TbRpiB despite being localized mainly in the cytosol can be also found in glycosomes due to colocalization with the glycosomal marker , aldolase [38] ( Fig . 2B ) . Upon digitonin fractionation , RpiB showed an intermediate pattern between the glycosomal marker , aldolase ( still partially in the pellet after 200 µg/ml digitonin treatment ) and the cytosolic marker , enolase ( almost all in supernatant with 25 µg/ml digitonin ) , being practically released with 100 µg/ml digitonin ( Fig . 2C ) . In conclusion , RpiB localizes mainly in the cytosol of bloodstream forms . To assess if TbRpiB targeting affects in vitro bloodstream forms growth , RNAi against RpiB was induced . This resulted in a lower mRNA and protein levels 1 and 2 days post-induction ( Fig . 3A and B , respectively ) . Using ImageJ software we estimate a decrease of approximately 93% of protein levels at 48 h RNAi post-induction . The growth of TbRpiB RNAi tet ( - ) and wt tet ( - ) cell lines was shown to be similar ( Fig . 3C ) . A significant decrease of in vitro cell proliferation of induced versus non-induced RNAi cell lines was seen only after day 4 of the cumulative growth curve ( Fig . 3C ) . To test the importance of RpiB for parasite infectivity in a disease model , two groups of BALB/c mice were inoculated with the wt parental cell line and other two groups with the RNAi cell line . Some mice were fed with water containing doxycycline ( Dox ) to induce downregulation of TbRpiB , whilst the remaining mice were kept as non-induced controls . A Western blot confirmed the reduction of the protein level in 48 h RNAi induced parasites used for mice infections ( Fig . 4A ) . Blood samples were taken from all mice at daily intervals to chart parasitaemia ( Fig . 4B ) . Animals achieving a parasitaemia greater than 108 trypanosomes per millilitre were euthanized . In vivo growth of the TbRpiB RNAi Dox ( - ) trypanosomes was not significantly different from that of wt Dox ( - ) parasites . However a significant decrease in the parasitaemia of induced versus non-induced RNAi cell lines was seen . Within 6 days of inoculation , contrary to mice infected with induced RNAi cell line ( in which overall parasitaemias remained below the detection limit , 5×104 trypanosomes/ml ) , mice infected with control parasites developed high levels of parasitaemia . As a consequence , and in contrast to mice infected with wt and TbRpiB RNAi Dox ( - ) parasites , which were culled sooner ( between eighth to thirteenth day post-infection ) , TbRpiB RNAi Dox ( + ) were euthanized from the eighteenth day post-infection ( Fig . 4C ) . Eventually parasitaemia also increased in the TbRpiB RNAi Dox ( + ) mice , due to the emergence of “RNAi revertants” ( Fig . 4D ) [39]–[42] . In this way , ribose 5-phosphate isomerase B despite being dispensable in vitro , confers optimal in vitro growth and is highly relevant for mice infections . Functional complementation of T . brucei RNAi cell lines with the T . cruzi homologue was performed , since TcRpiB has in vitro isomerase activity and TcRPIB nucleotide sequence is sufficiently different to avoid TbRpiB RNAi . Western blot analysis confirmed TbRpiB downregulation only in induced RNAi parasites , and TcRpiB expression exclusively in complemented parasites ( Fig . 5A ) . Cells with RNAi and complemented with TcRpiB grew equally in vitro ( Fig . 5B ) , and were almost as virulent in vivo ( Fig . 5C , D ) , as the wild-type . RNAi revertants appeared during the course of infection in induced TbRpiB RNAi infected mice , but not in induced complemented TbRpiB RNAi infected mice ( Fig . 5E ) . As a result , complementation restored in vitro and in vivo phenotypes .
In this study we demonstrated that TbRpiB , like the related TcRpiB and Leishmania donovani RpiB ( LdRpiB ) enzymes , has in vitro ribose 5-phosphate isomerase activity [31] , [43] . Based on the theoretical homology model , TbRpiB is predicted to be dimeric . Although the dimer comprises a complete functional unit , tetramers are observed in all available RpiB structures except that of Mycobacterium tuberculosis RpiB [36] . Similarly to T . cruzi , Clostridium thermocellum and Pisum sativum Rpi enzymes , TbRpiB has the ability of using both R5P or Ru5P as substrates , but with remarkable preference for Ru5P [31] , [44] , [45] . However , the differences in affinity are more pronounced in trypanosomes enzymes . Indeed , these differences were higher for TbRpiB compared to TcRpiB . Analysis of the three enzymes from trypanosomatids ( TcRpiB , LdRpiB and TbRpiB ) shows that TbRpiB and LdRpiB have the highest Km and kcat value for R5P substrate , respectively [31] , [43] . Nevertheless , we can speculate that such differences may result in part by the fact that parasite enzymes were expressed and purified as recombinant proteins in bacteria and not purified directly from trypanosomes extracts . Consequently , differences in protein post-transcriptional processing and/or changes in protein conformation cannot be excluded . RpiB is expressed on T . brucei procyclic and bloodstream forms , and our data indicate its higher expression in procyclics . Interestingly , a previous study has shown higher levels of TbRPIB mRNA ( Tb927 . 11 . 8970 ) in logarithmic phase procyclic forms compared to bloodstream forms [46] . However , its biological meaning , if any , remains to be elucidated . Regarding RpiB subcellular localization in bloodstream forms , the protein despite found mainly in the cytosol is also present in glycosomes . This might explain why a previous proteomic analysis failed to find TbRpiB enzyme in purified glycosomes [47] . The glycosomal localization observed within the dual-localization can be justified by the presence of a peroxisomal targeting signal , PTS2 ( -KVAIGADHI- ) , at the N-terminus [48] . Moreover , other enzymes of the hexose-monophosphate pathway , although present in glycosomes , were also found mainly within the cytosol ( e . g . glucose-6-phosphate dehydrogenase , 6-phosphogluconolactonase and transketolase ) [49] , [50] . TbRpiB is clearly needed for optimal in vitro parasite growth , although we do not know whether it is essential for survival since some protein remained after RNAi . Nevertheless , our results show that TbRpiB is important for parasites infectivity in vivo , through the appearance of RNAi revertants and reversion of the phenotype in complemented parasites . Infectivity defects of bloodstreams with reduced levels of TbRpiB were shown on a monomorphic T . brucei strain . This strain is abnormally virulent and typically mice do not survive longer than ≈10 days . In the future , it would be interesting to test the role of RpiB in a more chronic infection , as the one caused by pleomorphic strains . Interfering with the PPP non-oxidative branch showed to be detrimental under host pressure , in these highly proliferative parasitic forms , which can be due to a defective production of ribose 5-phosphate towards nucleotide and nucleic acid synthesis . Moreover , another enzyme capable of producing ribose 5-phosphate , ribokinase , is essential for parasites survival since attempts to remove the two alleles were unsuccessful [51] . TbRpiB is not the first protein reported as dispensable under standard laboratory culture conditions but crucial for parasites growth in the animal host [52] , [53] . In rich culture conditions , parasites may uptake essential nutrients from the extracellular medium , which may not be as available in blood . Moreover , in vivo , parasites need to deal with pressure from the host immune response . As for other proteins [54] , [55] , our in vitro results differ from the ones achieved in RNA interference target sequencing ( RITseq ) screen [56] . Indeed , proteins described to be significantly important for parasites fitness by Alsford and colleagues [56] were not in others studies [54] , [55] . Despite large-scale RNAi screens have already proved useful , caution should be taken due to some level of false negatives and positives , inherent to high-throughput approaches and more importantly due to off-target effects [57] . Furthermore , variations between different large-scale RNAi screenings were already been reported and explained by the use of different T . brucei strains , RNAi constructs and methods for assessing cell growth highlighting the importance of using complementary approaches in such studies [58] . Despite all , both studies are in agreement and show a role for TbRpiB on parasites growth . To further investigate if bloodstream forms deleted of RpiB are completely cleared in mice , studies with gene knockout parasites should be done . Overall our results clearly show a role of RpiB for bloodstream in vitro optimal growth and more importantly in vivo infectivity , but also suggest a conserved role among different Trypanosoma species . In conclusion TbRpiB emerges as a new potential therapeutic target against African sleeping sickness . | Within the non-oxidative branch of the pentose phosphate pathway , ribose 5-phosphate isomerase catalyzes the inter-conversion of ribose 5-phosphate and ribulose 5-phosphate . There are two types of ribose 5-phosphate isomerase , namely A and B . The presence of type B in Trypanosoma brucei , and its absence in humans , make this protein a promising drug target . African sleeping sickness is a serious parasitic disease that relies on limited chemotherapeutic options for control . In our study , a functional characterization of Trypanosoma brucei ribose 5-phosphate isomerase B is reported . Biochemical studies confirmed enzyme isomerase activity and its downregulation by RNAi affected mainly parasites infectivity in vivo . Overall this study shows that ribose 5-phosphate isomerase depletion is detrimental for parasites infectivity under host pressure . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"molecular",
"biology",
"biology",
"and",
"life",
"sciences",
"parasitology"
] | 2015 | Ribose 5-Phosphate Isomerase B Knockdown Compromises Trypanosoma brucei Bloodstream Form Infectivity |
In central neurons , the threshold for spike initiation can depend on the stimulus and varies between cells and between recording sites in a given cell , but it is unclear what mechanisms underlie this variability . Properties of ionic channels are likely to play a role in threshold modulation . We examined in models the influence of Na channel activation , inactivation , slow voltage-gated channels and synaptic conductances on spike threshold . We propose a threshold equation which quantifies the contribution of all these mechanisms . It provides an instantaneous time-varying value of the threshold , which applies to neurons with fluctuating inputs . We deduce a differential equation for the threshold , similar to the equations of gating variables in the Hodgkin-Huxley formalism , which describes how the spike threshold varies with the membrane potential , depending on channel properties . We find that spike threshold depends logarithmically on Na channel density , and that Na channel inactivation and K channels can dynamically modulate it in an adaptive way: the threshold increases with membrane potential and after every action potential . Our equation was validated with simulations of a previously published multicompartemental model of spike initiation . Finally , we observed that threshold variability in models depends crucially on the shape of the Na activation function near spike initiation ( about −55 mV ) , while its parameters are adjusted near half-activation voltage ( about −30 mV ) , which might explain why many models exhibit little threshold variability , contrary to experimental observations . We conclude that ionic channels can account for large variations in spike threshold .
Spike initiation in neurons follows the all-or-none principle: a stereotypical action potential is produced and propagated when the neuron is sufficiently excited , while no spike is initiated below that threshold . The value of that threshold sets the firing rate and determines the way neurons compute , for example their coincidence detection properties [1] , [2] . It is generally described as a voltage threshold: spikes are initiated when the neuron is depolarized above a critical value , when voltage-dependent sodium channels start to open . That biophysical mechanism is well understood since the studies of Hodgkin and Huxley in the squid giant axon [3] and subsequent modeling studies [4]–[7] . Recent findings have renewed the interest in the spike threshold . First , there is an intense ongoing debate about the origin of threshold variability observed in vivo [8]–[14] . In particular , it is unclear whether threshold variability is mainly due to experimental artifacts or molecular mechanisms , which might question the relevance of the Hodgkin-Huxley model for central neurons . Moreover , numerous experiments have shown that spike initiation does not only depend on the membrane potential but also on complex features of the inputs . For example , it depends on the preceding rate of depolarization [15]–[21] and on the preceding interspike intervals [12] , [22] . Those properties are functionally important because they enhance the selectivity of neurons in several sensory modalities , in particular in audition [23] , vision [24] , and touch [21] . Developmental and learning studies have also shown that the threshold adapts to slow changes in input characteristics . This phenomenon is known as long-term plasticity of intrinsic excitability and may be involved in the regulation of cell firing , short term memory and learning [25]–[31] . The excitability threshold also varies with the distance to the soma in a given neuron and with cell type [2] , [15] , [32]–[35] , which may explain functional differences . The modulation of cell excitability might be explained by the activation of voltage-gated potassium channel Kv1 [36]–[41] , inactivation of voltage-gated sodium channels [15] , [16] , [19] , [21] , fluctuations in sodium channel gating [42] , inhibitory synaptic conductance [43]–[45] and the site of spike initiation [14] . To understand the origin of spike threshold variability , we examined the role of several candidate mechanisms in biophysical neuron models: activation and inactivation of the sodium channel , slow voltage-gated channels ( e . g . Kv1 ) , synaptic conductances and the site of spike initiation . Our analysis is based on a simplification of the membrane equation near spike initiation and results in a simple formula for the spike threshold that quantifies the contribution of all those mechanisms . The threshold formula provides an instantaneous time-varying value which was found to agree well with numerical simulations of Hodgkin-Huxley type models driven by fluctuating inputs mimicking synaptic activity in vivo , and with simulations of a realistic multicompartmental model of spike initiation [54] .
To derive the threshold equation , we made a quasi-static approximation , assuming that all mechanisms that modulate the threshold are slow processes ( compared to the timescale of spike initiation ) . That threshold equation provides an instantaneous value of the spike threshold , as a function of modulating variables . Here we show how the dynamics of sodium inactivation , voltage-gated conductances and synaptic conductances translate into spike threshold dynamics .
Since Na channels are responsible for the generation of action potentials , the threshold is firstly determined by their activation characteristics . Activation curves for Na channels are well approximated by Boltzmann functions with similar slope factors ( ka = 4–8 mV in neuronal channels ) . The threshold is linearly related to the half-activation value Va and logarithmically related to the maximum Na conductance ga . The threshold also depends logarithmically on the Na inactivation variable h , so that it increases with the membrane potential and with every emitted spike . The modulating effect of inactivation is most pronounced when the half-activation value Vi is lowest ( i . e . , Na channels are partially inactivated at rest ) . Finally , the threshold depends logarithmically on the total conductance , which includes the leak conductance , voltage-gated conductances and synaptic conductances . In particular , Kv1 channels , which are expressed with high density at the spike initiation site [37] , [39] , [41] , increase the threshold in an adaptive manner ( the threshold increases with the membrane potential ) . This change in threshold occurs simultaneously with the effective membrane time constant , whereas threshold changes due to Na inactivation have no effect on the time constant , which might suggest a way to experimentally distinguish between the two effects . Indeed , the effective membrane time constant ( as measured in vivo for example in Léger et al . , 2003 [73] ) is ( C is the membrane capacitance ) while the effect of total conductance on spike threshold varies as , therefore as − . It is currently unclear whether threshold modulation is mainly due to Na inactivation or delayed-rectifier K currents . Our simulations with the multicompartmental model of spike initiation in pyramidal cells [54] suggest that the spike threshold is essentially determined by Na inactivation , but this may not be universally true . Recent experimental findings in hippocampal mossy fibers [74] suggest that delayed K+ currents are closed at spike initiation , which minimizes charge movements across the membrane and is thus more metabolically efficient . It emphasizes the fact that Na inactivation is a more metabolically efficient way to modulate spike threshold than K+ activation , since the former reduces charge transfer while the latter increases it . We have not considered the effect of channel noise , i . e . , fluctuations in Na channel gating [42] , [75]–[78] , which result in random threshold variations . Although dynamical equations of fluctuations in Na channel gating are well set [79] , [80] , they cannot be included in our theoretical framework because we neglected the time constant of Na activation ( which leads to the exponential model ) . There are two additional sources of variability which are artefactual: the fact that the threshold is not measured at the site of spike initiation , and threshold measurement methods . The latter source is difficult to avoid in vivo because only spike onsets can be measured . The former one also seems technically very difficult to avoid in vivo , since spikes are initiated in the axon hillock , which is only a few microns large . Although the soma and AIS are virtually isopotential below threshold , experimentally measured values of threshold differ between the two sites [34] because , as we previously remarked , in vivo measurements correspond to spike onset rather than threshold and therefore take place after spike initiation , when the two sites are not isopotential anymore . This experimental difficulty may introduce artefactual variability in threshold measurements [14] . To derive the threshold equation , we made several simplifying assumptions . First , we assumed that Na activation is instantaneous . It is indeed significantly faster than all other time constants but not instantaneous . The approximation is legitimate as long as the effective membrane time constant in the membrane equation is small ( , including all conductances ) , which is generally true before threshold . When Na channels open , the Na conductance dominates the total conductance and drastically reduces the effective time constant . Thus , we expect this approximation to be reasonable to predict spike initiation properties but not spike shape characteristics . Our second major assumption is a quasistatic approximation , i . e . , we assume that near spike initiation , all modulating variables and the input current can be considered as constant . In other words , we assume that the time constants ( except that of Na activation ) are larger than a few ms . This is clearly only a mathematically convenient approximation , but our predictions empirically agreed with numerical simulations . To investigate the role of Na inactivation , we also assumed that activation and inactivation are independent , which is a standard simplifying hypothesis ( Hille , 2001 ) . Although it is debatable [49] , [56] , it should remain valid in the case where activation and inactivation time constants are well separated . We also assumed that Na activation and inactivation curves were Boltzmann functions . Experimental data is indeed well fitted by Boltzmann functions , but the reported parameter values ( Va , ka ) correspond to fits on the entire voltage range , whereas we are interested in hyperpolarized voltage regions where the activation values are small . When only the relevant part of the experimental data is considered , different parameter values might be obtained . For example , when analyzing previously published biophysical models , we found that better results were obtained when Na activation curves , which were not exactly Boltzmann functions , were fitted in the spike initiation region ( −60 to −40 mV ) rather than on the entire voltage range ( Fig . 9 ) . We examined this issue in the biophysical model used in this paper ( see Materials and Methods ) . The Na activation curve of this model seemed to be well fit to a Boltzmann function ( Fig . 9A ) , however the fit was poor in the spike initiation zone ( −60 to −40 mV , Fig . 9C ) where activation is close to zero , which makes fit errors relatively larger . Although the slope factor is about 6 mV when the activation curve is fit over the entire voltage range , similar to experimental measurements [51] , it is only half this value when fit in the spike initiation region ( Fig . 9D ) , which explains why this model , as many other biophysical models , exhibits little threshold variability ( since threshold modulation is proportional to ) . We calculated the slope factor as a function of the voltage region and we found that it varies between 1 and 6 mV ( Fig . 9D ) . This finding motivates a reexamination of Na channel voltage-clamp data , focusing on the spike initiation region rather than on more depolarized regions , which are more relevant for spike shape than spike initiation . Fig . 10 addresses two potential difficulties . In experiments , activation curves are obtained by measuring the peak conductance after the clamp voltage is changed from an initial value V0 to a target value V , and normalizing over the entire range of target voltages . Thus , it assumes that inactivation is still at resting level h ( V0 ) when the peak current is measured . This would not be the case if the inactivation time constant τh were close to the activation time constant τm . Fig . 10A shows the effect of this overlap on the measurement of ka with simulated voltage-clamp data , where is a Boltzmann function with ka = 6 mV . It appears that ka is overestimated if τh is very close to τm , up to 50% when the two time constants are equal ( to 0 . 3 ms in these simulations ) . However the error quickly decreases as τh increases ( e . g . 10% error for τh = 1 ms ) . Another potential difficulty is the lack of data points in the relevant voltage range and the measurement noise , because currents are small . In Fig . 10B , we digitized an experimentally measured activation curve ( black dots ) , where clamp voltages were spaced by 5 mV . A Boltzmann fit over the entire voltage range gave ka = 7 . 2 mV while a fit over the hyperpolarized region V<−40 mV gave ka = 4 mV . However , the latter is not a reliable estimate because it corresponds to only 4 non-zero data points , which also seem to be corrupted by noise . Therefore it might be necessary to perform new measurements , specifically focusing on the spike initiation zone , perhaps with multiple measurements to reduce the measurement noise . Alternatively , ka could be measured with a phenomenological approach , using white noise injection in current clamp [22] . Another possible approach would be to directly fit the excitability model to the current-clamp response of a cell in which only Na channels would be expressed ( perhaps with fluctuating inputs ) . Finally , our analysis relies on a single compartment model . In the compartmental model , we found that between spikes , the membrane potential was 1 . 8±0 . 6 mV more depolarized at the soma than at the AIS . This is small compared to the slopes of all activation and inactivation curves in this model ( 5–9 mV ) . This agrees with our analysis of the electrotonic length in the subthreshold range , which is much larger than the distance between the soma and the AIS , although very fast synaptic inputs or proximal axonal inhibition could produce larger voltage gradients . Thus , our analysis should remain valid if the compartment represents both the soma and initiation site ( and also proximal dendrites ) . However , that approximation is not valid anymore after spike initiation ( see below ) . Spikes look sharper in the soma than in the AIS , presumably because they are initiated in the AIS and back-propagated to the soma [10] , [13] , [14] . That property is also seen in numerical simulations of multicompartmental models [34] , [70] . Yet , linear cable theory predicts the opposite property: the voltage at the soma is a low-pass filtered version of the voltage at the AIS , therefore spikes should look less sharp in the soma . Thus , increased sharpness must be due to active backpropagation of the action potential , which cannot be seen in a two compartment model ( such as described in Text S1 ) . From a theoretical point of view , the sharpening effect of backpropagation can be intuitively understood from the cable equation:It appears that the membrane equation is augmented by a diffusion term , which is positive and large in the rising phase of the action potential between the initiation site and the soma . Thus , for the same membrane potential V , the time derivative gets larger as this diffusion term increases , which sharpens action potentials . Sharpness can be measured in numerical simulations by plotting dV/dt vs . V in response to a suprathreshold DC current , and fitting it to an exponential model ( ) . In the model of Hu et al . [54] , we found that the slope factor , characterizing spike sharpness , was = 1 . 6 mV in the AIS and only 0 . 8 mV in the soma . This is in approximate agreement with empirical fits of exponential integrate-and-fire models to cortical neurons stimulated with fluctuating inputs , which reveal a surprisingly small slope factor , slightly above 1 mV [22] . Thus , in the multicompartmental model , active backpropagation did increase spike sharpness in the soma , but also in the AIS , since the slope factor was about twice smaller than predicted from fitting the Na activation curve to a Boltzmann function ( 3 . 6 mV ) . This increased sharpness did not affect the magnitude of threshold modulation . In single-compartment models , sharpness of spikes and threshold modulation are determined by the same quantity , related to the sharpness of the Na activation curve ( ka ) . It appears that this link does not hold anymore when active backpropagation is considered ( in multicompartmental models ) . Thus , in the threshold equation , the modulating factor is indeed ka ( from the Na activation curve ) rather than ( from spike sharpness , measured in the phase plot ( dV/dt , V ) ) . This explains that Na inactivation can produce large threshold variability ( 10 mV in our simulations ) even though spikes are very sharp .
We consider a single-compartment neuron model with voltage-gated sodium channels and other ion channels ( voltage-gated or synaptic ) , driven by a current I . The membrane potential V is governed by the membrane equation:where C is the membrane capacitance , gL ( resp . EL ) is the leak conductance ( resp . reversal potential ) , gi ( resp . Ei ) is the conductance ( resp . reversal potential ) of channel i , gNa ( resp . ENa ) the maximum conductance ( resp . reversal potential ) of sodium channels , PNa is the proportion of open Na channels and I is the input current . In this article , we used the following convention for conductances: lower case ( g ) for the total conductance over the surface of a compartment ( typically in units of nS ) and upper case ( G ) for conductances per unit area ( in units of S/cm2 ) . We assume that sodium channel activation and inactivation are independent , as in the Hodgkin-Huxley model [3] , i . e . , , where Pa is the probability that activation gates are open and Pi is the probability that a channel is inactivated . Following the Hodgkin-Huxley formalism , we define . The steady-state activation curve can be empirically described as a Boltzmann function [51]:where is the half-activation voltage ( ) and the activation slope factor ( ) . We make the approximation that Na activation is instantaneous and we replace Pa by its equilibrium value , so that . With instantaneous activation , the sodium current is:Action potentials are initiated well below Va ( about −30 mV , Angelino and Brenner , 2007 [51] ) , so that except during the spike . Similarly , ENa is very high ( about 55 mV ) , so that is not very variable below threshold . We make the approximation and we obtain:where . This approximation is meaningful for spike initiation but not for spike shape . With a reset ( ignoring inactivation and other ionic channels ) , we obtain the exponential integrate-and-fire model [48] , which predicts the response of cortical neurons to somatic injection with good accuracy , in terms of spike timings [22] , [81] , [82] . In this model , VT is the voltage threshold for constant input currents I and ka ( originally denoted ΔT ) is the slope factor , which measures the sharpness of spikes: in the limit mV , the model becomes a standard integrate-and-fire model with threshold VT ( although this is different in multicompartmental models , see Discussion ) . The resulting approximated model is thus:It is convenient to sum all conductances ( except for the Na channel ) , which gives a simpler expression:where is the total conductance and E* is the effective reversal potential:Finally , the inactivation variable h can be inserted in the exponential function:whereis the voltage threshold if all other variables are constant , i . e . , it is such that F′ ( ) = 0 , where F is the current-voltage function . The effect of Na inactivation on the threshold can be seen in the exponential model above , neglecting other conductances ( thus ) . Assuming that inactivation is slow compared to spike initiation ( quasi-static approximation ) , the voltage threshold is now , and it changes with the inactivation variable h . We assume , as in the Hodgkin-Huxley model , that inactivation has first-order kinetics:The steady-state value of the threshold is thus . We differentiate the threshold equation with respect to time:We now express h as a function of using the invert relationships: and :If the threshold remains close to its steady-state value ( ) , the equation simplifies to:with . The same method applies for voltage-gated conductances ( e . g . Kv1 ) . We compared our theoretical predictions with numerical simulations of a previously published point-conductance model with fluctuating synaptic inputs [83] . The membrane equation is:where and n are respectively the maximal conductance and the activation variable of thedelayed-rectifier potassium current , and and p are respectively the maximal conductance and the activation variable of the non-inactivating K current . All channel variables have standard Hodgkin-Huxley type dynamics . In Fig . 3A–C , only Na channel activation was considered , with instantaneous dynamics , i . e . , , h = 1 , n = 0 , p = 0 , I = 0:In Fig . 3D , the threshold equation was used to calculate VT for the Na channel properties reported in Angelino and Brenner ( 2007 ) [51] , since only the values of Va and ka were available . To evaluate our threshold equation with time-varying inputs ( Figs . 2C , 5 and 6 ) , we simulated the full conductance-based model with fluctuating synaptic conductances ( same parameters as in Destexhe et al . , 2001 [83] ) . In Fig . 6 , we shifted the voltage dependence of Na inactivation toward hyperpolarized potentials by −12 . 5 mV so as to obtain more threshold variability . To measure the time-varying threshold , we used a similar method as one previously used in vitro by Reyes and Fetz [84] , [61] . We simulated the model for 200 ms and measured the instantaneous threshold at regular time intervals T as follows . The model was simulated repeatedly with the same synaptic inputs ( frozen noise ) . In each trial , the neuron was depolarized at time nT ( only once per 100 ms run ) to a voltage value between −51 mV and −38 mV . With T = 0 . 6 ms and 65 voltage values , we ran 22 , 000 trials . The threshold at a given time is defined as the minimal voltage value above which a spike is elicited . The measured threshold was compared to the prediction obtained with the threshold equation ( see Results ) , where VT and ka were obtained from a Boltzmann fit to the activation function over the range −51 mV to −38 mV , giving VT = −68 mV and ka = 3 . 7 mV ( Va = −30 . 4 mV ) . The values of VT and ka depended on the fitting window ( see Discussion and Fig . 9 ) . In Fig . 7 , the voltage dependence of Na inactivation was shifted by −20 mV to induce more threshold variability ( giving Vi = −62 mV instead of −42 mV with the original parameter values ) and the maximum Na conductance was multiplied by 3 ( to keep threshold values in the same range ) . The standard deviations of synaptic conductances were also increased . In Fig . 8 , we simulated a multicompartmental model of spike initiation recently published by Hu et al . ( 2009 ) [54] , with fluctuating injected current modeled as an Ornstein-Uhlenbeck process ( mean 0 . 7 nA , standard deviation 0 . 2 nA , time constant 10 ms ) . The model was otherwise unchanged . The spike threshold , both at the soma and AIS , is defined at the voltage value when dV/dt first exceeds 10 V . s−1 preceding a spike . In some panels ( Fig . 8C , D , H ) , we extracted spikes from voltage traces by removing parts between spike onsets and 7 ms later . We estimated the activation properties of the Nav1 . 6 channel , which is responsible for spike initiation in this model , by fitting a Boltzmann function to the activation curve ( ) in the spike initiation zone ( −60 mV to −40 mV ) , which gave Va = −33 mV and ka = 3 . 6 mV . We then calculated the total maximal conductance of the Nav1 . 6 channel over the AIS , by integrating the channel density over the surface of the AIS ( using the morphology and channel density implemented in the published model code ) . We found gNa = 236 nS . Calculating the total leak conductance in this way was more difficult because leak channels were uniformly distributed on the whole morphology , including the dendrites , so that spatial attenuation should be taken into account . While this is theoretically possible using linear cable theory , we chose a simpler approach by directly measuring the membrane resistance at the soma with a DC current injection , and we found gL = 38 nS . With these values , the threshold equation predicted that the base threshold is VT = −55 . 9 mV . The model had a slow K+ current ( Im ) with the same channel density as the leak channels . Therefore the maximum total conductance was estimated as gKm = gL = 38 nS . It also had a fast K+ current which was distributed inhomogeneously on the whole neuron morphology , including dendrites . We estimated its total maximum conductance as , where the effective dendritic area was estimated from the ratio of total leak conductance over leak channel density , i . e . , . We found gKv = 906 nS . We then calculated the theoretical threshold using these parameters and the instantaneous values of the relevant channel variables ( h , nKm , nKv4 ) . In Fig . 10A , we simulated a voltage clamp experiment in a simplified model with only Na channels , assuming the leak current was subtracted , where both activation and inactivation curves ( and ) were Boltzmann functions , with parameters Va = −30 mV , ka = 6 mV , Vi = −65 mV and ki = 6 mV . The activation and inactivation time constant were fixed ( τm = 0 . 3 ms and τh between 0 . 3 and 3 ms ) . The conductance was measured at the current peak after the clamp voltage was switched from a fixed initial voltage V0 = −70 mV to a test voltage V , which was varied between −100 and 50 mV ( the current was divided by V-ENa to obtain the conductance , and we assumed that ENa was known - in an experiment it would be obtained from a linear fit to the highest voltage region ) . The conductance was normalized by the maximal conductance over the tested voltage range and the resulting curve was fit to a Boltzmann function in the hyperpolarized region V<−40 mV . All simulations were written with the Brian simulator [85] on a standard desktop PC , except the simulation of the multicompartmental model of Hu et al . [54] , for which we used Neuron . | Neurons communicate primarily with stereotypical electrical impulses , action potentials , which are fired when a threshold level of excitation is reached . This threshold varies between cells and over time as a function of previous stimulations , which has major functional implications on the integrative properties of neurons . Ionic channels are thought to play a central role in this modulation but the precise relationship between their properties and the threshold is unclear . We examined this relationship in biophysical models and derived a formula which quantifies the contribution of various mechanisms . The originality of our approach is that it provides an instantaneous time-varying value for the threshold , which applies to the highly fluctuating regimes characterizing neurons in vivo . In particular , two known ionic mechanisms were found to make the threshold adapt to the membrane potential , thus providing the cell with a form of gain control . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"neuroscience/theoretical",
"neuroscience",
"computational",
"biology/computational",
"neuroscience"
] | 2010 | A Threshold Equation for Action Potential Initiation |
Knowledge of the Free Energy Landscape topology is the essential key to understanding many biochemical processes . The determination of the conformers of a protein and their basins of attraction takes a central role for studying molecular isomerization reactions . In this work , we present a novel framework to unveil the features of a Free Energy Landscape answering questions such as how many meta-stable conformers there are , what the hierarchical relationship among them is , or what the structure and kinetics of the transition paths are . Exploring the landscape by molecular dynamics simulations , the microscopic data of the trajectory are encoded into a Conformational Markov Network . The structure of this graph reveals the regions of the conformational space corresponding to the basins of attraction . In addition , handling the Conformational Markov Network , relevant kinetic magnitudes as dwell times and rate constants , or hierarchical relationships among basins , completes the global picture of the landscape . We show the power of the analysis studying a toy model of a funnel-like potential and computing efficiently the conformers of a short peptide , dialanine , paving the way to a systematic study of the Free Energy Landscape in large peptides .
Polymers and , more specifically , proteins , show complex behavior at the cellular system level , e . g . in protein-protein interaction networks [1] , and also at the individual level , where proteins show a large degree of multistability: a single protein can fold in different conformational states [2]–[4] . As a complex system [5] , [6] , the dynamics of a protein cannot be understood by studying its parts in isolation , instead , the system must be analyzed as a whole . Tools able to represent and handle the information of the entire picture of a complex system are thus necessary . Complex network theory [7] , [8] has proved to be a powerful tool used in seemingly different biologically-related fields such as the study of metabolic reactions , ecological and food webs , genetic regulatory systems and the study of protein dynamics [7] . In this latter context , diverse studies have analyzed the conformational space of polymers and proteins making use of network representations [9]–[12] , where nodes account of polymer conformations . Additionally , some studies have tried to determine the common and general properties of these conformational networks [13] , [14] looking at magnitudes such as clustering coefficient , cyclomatic number , connectivity , etc . Recently , trying to decompose the network in modules corresponding to the free energy basins , the use of community algorithms over these conformational networks have been proposed [15] . Although this approach has opened a promising path for the analysis of Free Energy Landscapes ( FEL ) , the community based description of the network leads to multiple characterizations of the FEL and thus it is difficult to establish a clear map from the communities found to the basins of the FEL . A similar approach , commonly used to analyze the complex dynamics , is the construction of Markovian models . Markovian state models let us treat the information of one or several trajectories of molecular dynamics ( MD ) as a set of conformations with certain transition probabilities among them [9] , [16] , [17] . Therefore , the time-continuous trajectory turns into a transition matrix , offering global observables as relaxation times and modes . In [16]–[18] the use of Markovian models is proposed with the aim of detecting FEL meta-stable states . However , the above approaches to analyze FELs of peptides involves extremely large computational cost: either general community algorithms or large transition matrices . Finally , other strategies to characterize the FEL that have successfully helped to understand the physics of biopolymers , are based on the study of the Potential Energy Surface ( PES ) [3] , [4] , [19]–[21] . The classical transition-state theory [22] allows us to project the behavior of the system at certain temperature from the knowledge of the minima and transition states of the PES . This approach entails some feasible approximations , such as harmonic approximation to the PES , limit of high damping , assumption of high barriers , etc . These approximations could be avoided working directly from the MD data . In this article we make a novel study of the FEL capturing its mesoscopic structure and hence characterizing conformational states and the transitions between them . Inspired by the approaches presented in [12] , [15] and [16] , [17] , we translate a dynamical trajectory obtained by MD simulations into a Conformational Markov Network . We show how to efficiently handle the graph to obtain , through its topology , the main features of the landscape: conformers and their basins of attraction , dwell times , rate constants between the conformational states detected and a coarse-grained picture of the FEL . The framework is shown and validated analyzing a synthetic funnel-like potential . After this , the terminally blocked alanine peptide ( Ace-Ala-Nme ) is studied unveiling the main characteristics of its FEL .
First , we encode a trajectory of a stochastic MD simulation into a network: the CMN . This map will allow us to use the tools introduced henceforth to analyze a specific dynamics of complex systems such as biopolymers . Up to now , we have illustrated the conversion of molecular dynamics data into a graph ( the CMN ) . Now , we show how to efficiently obtain the thermo-statistical data from the mesoscopic description of the CMN .
The alanine dipeptide , or terminally blocked alanine peptide ( Ace-Ala-Nme , Figure 3A ) , is the most simple “biological molecule” that exhibits the common features shown by larger biomolecules . Despite of its simplicity , this system has more than one long-life conformational state with different transition pathways . Since the first attempt by Rossky and Karplus [37] to model this dipeptide solvated , this system has been widely studied in theoretical works [38]–[41] . The alanine dipeptide has been also the appropriate molecule to test tools to explore the FEL [15] , [16] , [42] and , specifically , to study reaction coordinates [41] , [43] . The alanine dipeptide has two slow degrees of freedom , the rotatable bonds ( ) and ( ) ( see Figure 3A ) . The FEL projected onto these dihedral angles let us identify the conformational states that characterize the geometry of biopolymers , namely: alpha helix right-handed ( ) , alpha helix left-handed ( ) , beta strands ( , ) , etc . The number of local minima in the ( , ) space depends on the effective potential model used to simulate the system . Up to date , electronic structure calculations have identified a total of nine different conformers [44] . Regarding MD simulations different conformational states have been observed: ( i ) using classical force fields with explicit solvent up to six conformers are detected [16] , [38] , [39] , ( ii ) at least four stable states by using implicit solvent [15] , [38] , [40] , and ( iii ) two stable conformers in vacuum conditions [38] , [41] . On the other hand , since the angles and seem appropriate to distinguish the metastable states , the kinetics between them is not accurately described with this choice of reaction coordinates , the solvent coordinates and/or other internal degrees of freedom must be taken into account [41] , [43] . We have used SSD algorithm to detect the local minima and their corresponding basins for this molecule in the space . For this purpose , a Langevin MD simulation of 250 ns has been performed at a temperature of 400 K ( see Text S3 for further details ) . Additionally the CMN has been built dividing the Ramachandran plot into cells of surface 9°×9° ( 40×40 ) and taking dialanine conformations at time intervals of . The resulting CMN have a total of nodes and directed links . The algorithm applied to the CMN network reveals 6 basins . Figure 3B shows the resulting network where nodes belonging to the same basin take the same color . Bringing back this information to the Ramachandran map , these 6 sets of nodes define 6 regions represented in Figure 3C . To better illustrate this division , other representation , where each region has a different color , is shown . By comparing with previous studies on this molecule , we identify the regions in orange , red , yellow and pink with conformers , , and respectively [38] , [45] . Besides , region green corresponds to conformer and the blue one to [16] , [45] . Remarkably , the basin ( one of the less populated state ) has been visited 1155 times with a mean stay time of 4 . 20 ps . We now look at the coarse-grained picture of the FEL by describing the properties of the 6 basins detected . The different weights of the basins are related to the free energy of the corresponding conformational macro-states . In Table 1 these energy differences are shown , taking the most populated basin as the energy reference . The lowest free energy basins correspond to configurations with φ≤0° ( , , and ) , whereas the two other conformers located in the region φ≥0° have the highest free energy but the largest dwell time . Moreover , we have also analyzed the trapping efficiency of each basin by reporting the mean escape time ( ) as well in Table 1 . The FEL can be represented as a dendogram , see Figure 4 , where the hierarchical map of the conformers based on Free Energy gives at first glance a global picture of the landscape . Remarkably , the conformer , despite of having one of the highest free energy , looks like the metastable state with longest life . This result is supported by the values of Mean Escape Time shown in Table 1 . The alanine dipeptide has been also studied because of its “fast” isomerization and a “slow” transition . Our coarse-grained picture of the FEL also allows us to extract information about these transitions . In Table 2 we show some of the relevant characteristic transition times from a basin a to an adjacent basin b , . [The whole data is shown in the Text S3 . ] Transitions between basins with the same sign of are remarkably faster ( e . g and ) . While slow transitions are observed for those hops crossing the line φ = 0° ( and ) , showing them as rare events . Instead , the alanine dipeptide finds more easily paths to go to φ≥0° conformers through and by crossing φ = 180° . To round off the description of the FEL , the dendogram corresponding to the temporal hierarchy is shown in Figure 5 . From the figure , it becomes clear that the behavior of the dialanine depends on the time scale used for its observation . Again , the same two different sets of conformers are distinguished from this hierarchy . Additionally , the global minimum conformer is reached in around 100 ps from any basin . Finally , the magnitudes computed here for the alanine dipeptide would allow to construct a first-order kinetic model of 6 coupled differential equations as Eq . ( 6 ) ( assuming equilibrium intra-basin ) . This model contains the same information as the kinetic model by Chekmarev et al . for the irreversible transfer of population from [40] .
Hierarchical landscapes characterize the dynamical behavior of proteins , which in turn depends on the relation between the topology of the basins , their transitions paths and the kinetics over energy barriers . The CMN analysis of trajectories generated by MD simulations is a powerful tool to explore complex FELs . In this article , we have proposed how to deal with a CMN to unveil the structure of the FEL in a straightforward way and with a remarkable efficiency . The analysis presented here is based on the physical concept of basin of attraction , making possible the study of the conformational structure of peptides and the complete characterization of its kinetics . Note that this has been done without the estimation of the volume of each conformational macro-state in the coordinates space and without the ‘a priori’ knowledge of the saddle points or the transition paths from a local minimum to another . On the other hand , the framework introduced in the article provides us with a quantitative description of the dialanine's FEL , coming up directly from a MD dynamics at certain temperature . The peptide explores its landscape building the corresponding CMN and the success of extracting the relevant information is up to the ability of dealing with it . Neither the FE basins were defined by the unique criterion of clustering conformations with a geometrical distance [46] , nor the rate constants were projected from the potential energy surface [19] , [20] . Moreover , the conformers and their properties were computed from the MD with the only limitation of the discretization of time and space . Although we have applied the method to low dimensional landscapes , we expect that high dimensional systems could be also studied , by the combination of this technique with the usual methods to reduce the effective degrees of freedom ( like principal component analysis or essential dynamics ) . In conclusion , the large amount of information obtained by working with the CMN , its potential application to any peptide with a large number of monomers , and the possibility of performing the analysis on top of CMN constructed via several short MD simulations [47] , make the approach presented here a promising way to describe the FEL of a protein . | A complete description of complex polymers , such as proteins , includes information about their structure and their dynamics . In particular it is of utmost importance to answer the following questions: What are the structural conformations possible ? Is there any relevant hierarchy among these conformers ? What are the transition paths between them ? These and other questions can be addressed by analyzing in an efficient way the Free Energy Landscape of the system . With this knowledge , several problems about biomolecular reactions ( such as enzymatic activity , protein folding , protein deposition diseases , etc . ) can be tackled . In this article we show how to efficiently describe the Free Energy Landscape for small and large peptides . By mapping the trajectories of molecular dynamics simulations into a graph ( the Conformational Markov Network ) and unveiling its structural organization , we obtain a coarse grained description of the protein dynamics across the Free Energy Landscape in terms of the relevant kinetic magnitudes of the system . Therefore , we show the way to bridge the gap between the microscopic dynamics and the macroscopic kinetics by means of a mesoscopic description of the associated Conformational Markov Network . Along this path the compromise between the physical nature of the process and the magnitudes that characterize the network is carefully kept to assure the reliability of the results shown . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"computational",
"biology/molecular",
"dynamics",
"computational",
"biology/macromolecular",
"structure",
"analysis",
"biophysics/theory",
"and",
"simulation",
"biophysics/protein",
"folding"
] | 2009 | Exploring the Free Energy Landscape: From Dynamics to Networks and Back |
Leprosy Type 1 reactions are difficult to treat and only 70% of patients respond to steroid treatment . Azathioprine has been used as an immune-suppressant and we tested its efficacy in treating leprosy T1R . Randomised controlled trial adding azathioprine to steroid treatment for leprosy reactions . This trial was conducted in four leprosy hospitals in India . Patients with a new leprosy Type 1 reaction affecting either skin or nerve were recruited . They were given a 20 week course of oral prednisolone either with placebo or azathioprine 50mg for 24 , 36 or 48 weeks . Outcomes were measured using a verified combined clinical reaction severity score ( CCS ) and the score difference between baseline and end of study calculated . An intention to treat analysis was done on the 279 patients who had an outcome . 345 patients were recruited , 145 were lost due to adverse events , loss to follow up or death . 36% needed extra steroids due to a recurrence of their skin and/or nerve reaction . 76% of patients had improvements in their CCS the end of the study , 22% had no change and 1 . 1% deteriorated . Adding azathioprine to steroid treatment did not improve CCS . So the improvements were attributable to treatment with steroids . We analysed the skin , sensory and motor scores separately and found that skin improvement contributed most with 78 . 9% of patients having skin improvement , azathioprine treatment for 48 weeks improved sensory scores it also improved motor scores but so did treatment with prednisolone alone . We identified significant adverse effects attributable to steroid treatment . When azathioprine and Dapsone were given together significant numbers of patients developed significant anaemia . Azathioprine is not recommended for the treatment of leprosy reactions and does not improve steroid treatment . Recurrent reactions are a major challenge . We have also identified that 65% of patients with sensory and 50% with motor nerve damage do not improve . Future studies should test giving azathioprine in the treatment of nerve damage and giving a higher dose for 48 weeks to patients . These findings highlight the difficulty in switching off leprosy inflammation and the need for better treatments for reactions and nerve damage . There is also a research need to identify patients who have recurrences and optimize treatments for them . Patients with recurrences may benefit from combined treatment with steroids and azathioprine . We have also shown that significant numbers of patients treated with steroids develop adverse effects and this needs to be highlighted in leprosy programmes . Research is needed to identify patients who do not respond to steroid treatment and develop alternative treatments for them . ClinicalTrials . gov This trial was registered with the Indian Council of Medical research clinical Trial register as a clinical trial Number—REFCTRI/2016/12/007558
Leprosy is complicated by immune–mediated reactions which affect about 30% of patients with Borderline leprosy . [1] These manifest clinically with inflammation of skin lesions and neuritis ( nerve inflammation ) which produces loss of sensory and motor nerve function . They are important because the loss of nerve function can lead to long term disability and the inflamed skin lesions are disfiguring . The underlying pathology of these inflammatory episodes is increased delayed type hypersensitivity to M leprae antigens with CD4 cells in lesions and activated macrophages and oedema . T1Rs are mediated via Th1 type cells and the pro-inflammatory cytokines , IL12 TGF b and the oxygen-free radical producer iNOS are all found in lesions . [2 , 3] Reactions may occur before , during or after anti-bacterial Multi Drug Therapy . There are few good randomised controlled trials on the effectiveness of steroids in the treatment of T1R , and a Cochrane Review only included three trials . [4] Steroids are widely used in leprosy reactions and most skin lesions improve with treatment but nerve function improvement is less satisfactory , with a median improvement rate of 50% ( Range 30–86% ) . An RCT done in India comparing three different regimens found better outcomes with 20 versus 12 weeks of treatment , whereas the outcomes with different starting doses 60 mg or 30 mg were similar . [5] Alternatives to steroid treatment are needed because of the high relapse rate for T1R [6] and patients having contraindications to steroid treatment such as diabetes , an increasing problem in India . Alternative treatments are also needed for patients with reactions who do not improve on steroid treatment and those who become steroid dependent . Since the underlying pathology is one of immune mediated inflammation we hypothesised that adding in additional immuno-suppression would improve outcomes . Azathioprine is an immuno suppressant used in immune mediated diseases . [7] It is metabolized to mercaptopurine by the enzyme thiopurine methyltransferase which inhibits T cell development . A trial in Nepalese leprosy patients with T1R compared a 12 week Azathioprine Prednisolone combination versus prednisolone alone and found the dose of prednisolone needed was less in those on the combined regimen . [8] One of the shortcomings of the Nepal study was that the azathioprine was only used for 12 weeks which is insufficient time to assess clinical efficacy . We therefore wanted to test the efficacy of azathioprine over a longer period . Adverse effects due to azathioprine may occur in up to 15% of patients and include nausea , vomiting and mucosal ulcers and bone marrow suppression . [9] Careful clinical monitoring of adverse effects of azathioprine is needed , especially in resource poor settings . We tested our hypothesis that azathioprine would be more efficacious than prednisolone alone by conducting a four arm randomised controlled trial . This study was done in four leprosy hospitals in north India; patients with T1R and/or neuritis were recruited and given a 48 week course of treatment . This comprised either a 20 week course of prednisolone alone , or prednisolone plus azathioprine for either 24 , 36 or 48 weeks in a double blind design . The objectives of the trial were therefore to determine whether the addition of Azathioprine to Prednisolone The outcome measures were skin inflammation measures and neurological measures which assess sensory and motor function in peripheral nerves . We used scales for assessing skin lesions developed for a previous trial on T1R . [10]
Four TLM ( The Leprosy Mission ) centres participated in this study: The Leprosy Mission Community Hospital , Shahdara , New Delhi; The Leprosy Mission Hospital , Faizabad , Uttar Pradesh; The Leprosy Mission Hospital Champa , Chattisgarh; and Purulia Leprosy Home and Hospital , West Bengal . The trial started in Aug 2007 and finished in 2012 . All patients received a 20 week course of prednisolone ( P ) as follows: P was started at a dose of 40 mg per day and reduced by 5 mg/day every two weeks till a dose of 20 mg/day was reached; 20 mg/day was given for four weeks , then 15 mg/day for four weeks . After this 10 mg/day P was given for two weeks and 5 mg/day for two weeks , ( 40x2 , 35x2 , 30x2 , 25x2 , 20x4 , 15x4 , 10x2 and 5x2 = 20 weeks ) . Patients were randomised to receive concomitant medication for 48 weeks with azathioprine ( A ) 50 mg fixed dose for 24 , 36 or 48 weeks ( APC ) or a placebo . Each active treatment arm was made up to 48 weeks with placebo for 28 , 24 , or 12 weeks , so all patients took 48 weeks of medication . The medication was prepared by and stored at The Leprosy Mission Research Centre , Delhi . Numbered treatment packs were made up and delivered to the centres . Specially designed forms were used for each stage of the recruitment , consent , assessment , laboratory investigations and follow-up process . The following patient based definitions were used Paucibacillary ( PB ) leprosy—when a patient had five or fewer well-demarcated , hypo-pigmented and anaesthetic skin lesions , and/or one peripheral nerve thickened . Multi bacillary ( MB ) leprosy—when a patient had more than five skin lesions and/or more than one nerve involved . Skin Type 1 Reaction—when patient had erythema and oedema of skin lesions . Nerve Function Impairment ( NFI ) —present when a patient had new sensory impairment , motor impairment or both . Nerve pain alone was not included in the definition of neuritis . New nerve function loss–is ‘New’ when it occurred within the previous six months . This was established by asking the patient about symptoms and duration of nerve function loss . Old nerve function loss—is ‘Old’ when it started six months or more before the date of presentation . This was established by asking the patient about symptoms and duration of nerve function loss . Motor function loss—is present when there was a muscle score below five in any of the muscles on voluntary muscle testing . The nerves tested were facial , ulnar , median , radial and lateral popliteal with the following muscles/movement strengths: orbicularis oculi , abductor digitii minimi , abductor pollicis brevis , wrist extension , tibialis anterior . Sensory function loss—is defined as a loss of two or more sensory points on Semmes Weinstein testing in the area supplied by a sensory nerve . Sensory testing was done for ulnar , median and posterior tibial nerves . Worsening of motor function–is defined as a change in one or more than one grade in the voluntary muscle testing . Worsening of sensory function—is defined as a change in two or more than two points in the sensory testing . Nerve recovery—is when the NFI scores regain normal values for either sensory or motor functions . Nerve improvement—is when the NFI scores improve from the baseline value for either sensory or motor functions but does not regain normal values . Most severely affected nerve—is a nerve with a score of 3 on motor testing and /or 1 . 5 on sensory testing . Least affected nerve—is a nerve with a score of 1 on motor testing and/or 0 . 5 on sensory testing . Recurrent reactional skin lesions—are defined as newly erythematous lesions that have either developed whilst on steroids or when patient has stopped steroids . Restarting steroids–is restarting or increasing steroid therapy during the study in the presence of worsening nerve function or recurrent skin reactions as defined above . Defaulter–is a patient who did not return to the hospital within four weeks of the last visit and so is removed from the study . Withdrawn–is when a patient is removed from the study due to adverse effects . Improved combined score- measured by skin , motor and sensory scores . Sensory testing ( ST ) was performed using Semmes-Weinstein monofilaments ( SWM ) ( Sorri-Bauru , Bauru , São Paulo , Brazil ) at designated test sites on the hands and feet . [11] The sensation in the areas of skin supplied by the ulnar and median nerves was tested with 2g SWM at six sites . Sensation on the feet was tested using 10g SWM at four sites . For both sets of sensory testing a score was developed . For each nerve the score ranged from 0–1 . 5 depending on the number of sites at which the monofilament was perceived . The range of scores for each foot was ( 0–1 . 5 ) . The sensory scores were summated on every visit . ( See Appendix 1 ) . A similar method of assessment was used in a published study . [6]A score of 0 for an individual nerve indicated that complete sensation was present and a score of 1 . 5 that there was no sensation in the area supplied by that nerve . A patient with a sensory score of 3 either had slight impairment at several points in different nerves or two nerves with complete loss of sensation . A modification of the Walker scale was needed because the two studies used different sets of monofilament for sensory testing . In the AZALEP study 2 and 10 gram monofilaments were used on the hands and feet respectively whereas in the Walker study 2 and 10 gram monofilaments were used on the hand , 10 and 300 gram monofilaments were used on the feet . Voluntary muscle testing ( VMT ) was assessed using the modified Medical Research Council ( MRC ) grading of power . The grading was modified in our scores: a MRC grade 5 = 0 , 4 = 1 , 3 = 2 and 2 , 1 and 0 were grouped together as 3 . The scores range from 0–3 for each muscle with a total range of 0–30 . The scores for each muscle were summated on each visit . A score of 0 indicated completely intact muscle power and 30 would indicate severe weakness in all the muscles tested . A score of 9 could indicate loss of function in three muscles or slight weakness in all muscles ( See supporting Information ) . The skin , sensory and motor scores were summated for each patient on each visit . Trial outcomes were assessed by looking at score differences . For each individual nerve the presence of new nerve damage was determined from the patient’s history and VMT/ST . New nerve damage had to be present for the patient to be entered into the study . Old nerve damage was separately recorded because it was not expected that these nerves would improve . Patients with PB and MB leprosy were included if they had Patients already on steroid therapy ( started within the past four weeks ) were included if they presented with new neuritis . Patients taking or having completed MDT were eligible . The following exclusion criteria ( all defined in the protocol ) were applied: age less than 15 years or more than 60 years , weight less than 30 kg , confirmed pregnancy , on treatment for tuberculosis , known HIV sero-positivity , hepatic dysfunction , bone marrow dysfunction , splenomegaly , hypertension and diabetes , Erythema Nodosum Leprosum ( ENL ) reactions . Patients unable to comply with monitoring and follow up requirements were also excluded . All patients had a full blood count ( haemoglobin , white cell , differential and platelet counts ) , renal function test ( serum creatinine ) , liver function tests ( SGOT & SGPT ) and a random blood sugar done to detect underlying conditions for exclusion . Ethical approval for this study was obtained from both the London School of Hygiene & Tropical Medicine , London UK and the Ethics Committee of The Leprosy Mission Trust India on 8th February 2008 . This trial was registered with the Indian Council of Medical research clinical Trial register as a clinical trial Number—REFCTRI/2016/12/007558 Informed Consent was taken from patients who were eligible to be included in the study . All eligible patients were invited to participate in the study by the study Medical Officer ( MO ) . They received a detailed information sheet ( Form 1 . 0 ) about the Azathioprine study ( in the local language ) ; a patient counsellor helped them understand the study details . The patients then had the above investigations . Those found eligible returned to the study MO , where they gave their signed informed consent ( Form 2 . 0 ) to participate in the study , and were included . Patients who were either ineligible for trial entry or who did not wish to participate were treated as per the routine practice in that hospital . Patients were also informed that if they did not consent , this would not affect their usual treatment . The sample size calculation was based on the expected improvements in the combined scores ( skin , sensory & motor ) that the addition of azathioprine to prednisolone would produce . It was calculated that the addition of azathioprine would produce a 25% improvement in the combined scores . This would be a score of 4 allowing for an alpha of 0 . 05 and beta of 0 . 80 , the minimum sample size is calculated as follows: n=2x[ ( Zα+Z1-β ) xSD ) /d]2=2x[ ( 1 . 96+0 . 842 ) ) x2 ) /1]2=63pergroup . We assumed a ‘lost to follow up’ rate of 15% , and dropout rate due to adverse reactions of 10% , so minimum of 78 patients would be required per arm , and a total of 312 . The staff at the centres were blinded to the randomisation . The Head of the Research Resource Centre ( RRC ) , using Random Sampling Numbers prepared the confidential allocation into four arms , which was balanced for each centre , and created lists with an unique serial number . The allocation for each patient was kept confidential at the RRC . The packing was done at RRC in small daily pill boxes with the appropriate amounts of each drug for each arm for four weeks . The boxes were prepared for 48 weeks for each patient . Thus there were 12 boxes for each patient , distributed to the respective centres by courier . The drugs were stored in cool dry rooms both at RRC and at the centres . Drugs were ordered in batches to ensure that they were within the active period after manufacture . The placebo tablet was identical to the Azathioprine tablet . Azathioprine Tablets BP 50 Mg ( TRANSIMUNE* ) and the placebo tablets were manufactured by Troikaa Pharmaceuticals Ltd . Thol-382 728 , Gujarat , India . Prednisolone Tablets I . P were manufactured by Comprehensive Medical Services India , ( Essential Drugs Project ) Riverside , P . B . No . 988 , Manapakkam , Chennai -600 089 . Physiotherapists in each hospital identified new eligible patients . The patient had clinical and laboratory investigations according to the flow chart ( Appendix 2 ) and the clinical history with details of skin type 1 reaction , nerve function impairments and other clinical details was recorded . The pre-treatment assessment form was filled in; they were given a box with four weeks treatment and a date for the follow up visit . Patients were followed up fortnightly for the first eight weeks and then at four week intervals , with full clinical and neurological assessments and laboratory investigations done at each visit . If the scores had worsened , the patients were observed for seven days , with repeat testing and the protocol for increasing prednisolone was followed . Compliance with the medication was recorded . Possible adverse effects were detected by relevant questioning and investigations . Any participant who did not return for a follow up for more than two weeks was contacted at his / her home by phone or personal visit and invited to return . Any patient who returned to the hospital within four weeks of the last visit continued in the study and those who did not return were classed as defaulters . In patients who developed infections during the first 20 weeks , their dose of prednisolone was tapered more quickly ( decreasing by 5mg every three days ) and stopped , and antibiotic therapy relevant to the presumed infection given . In patients who were on MDT and developed anemia , dapsone was stopped . If the adverse events resolved within four weeks , the patient was restarted on the drugs and continued in the study . If this period exceeded four weeks or the adverse event was severe the patient was withdrawn from the study . If the neuritis worsened or there was a recurrence of type 1 skin reaction during the first 20 weeks , the dose of prednisolone was increased to the preceding dose resulting in a longer duration of steroid therapy . If worsening occurred after 20 weeks , when prednisolone treatment had finished , the patient was given a 12 week course of prednisolone , starting with 40mg as in the protocol ( two weeks each of 40 , 30 , 20 , 15 , 10 & 5 mg of prednisolone ) . Potential adverse effects of steroids , azathioprine and MDT were closely monitored . Protocols were developed for the investigation and management of all complications . Adverse effects and deaths were discussed by the clinical team and ascribed when possible to one of the drugs . Deaths were investigated and the cause of death established by the MO . The contribution of Azathioprine and prednisolone to the cause of death were reviewed . All the patients’ details were recorded in predesigned forms designed for the study , at baseline and at each subsequent visit . These covered Screening for Eligibility , Registration for Trial , Informed Consent , Baseline History & Examination , Baseline Nerve Function Measurement , Baseline Lab Investigations . Data from the clinical trial including the baseline data were computerised using dedicated software as well as on MS Excel sheets . Data was transferred into Statistical Package for Social Sciences and analysed . The trial data is stored at the Leprosy Research Centre , Delhi . It can be obtained from the Head of the Leprosy Mission Research Centre , India , currently Dr annamma John . Chi-square testing was used to compare the variables such as age , sex , type of leprosy . For each of the primary and secondary outcomes , the differences between the baseline and end of study were computed , for each arm and the paired t-test was used to determine the statistical significance . The actual p-value was noted and inferences made accordingly using the conventional 5% and 1% levels . The differences between the patients receiving placebo and each of the three Azathioprine treatment arms were also tested for statistical significance using the independent t-test for the various outcomes . A modified intention to treat analysis was done on the outcomes for all the patients entered in the study ( 345 ) excluding those who had an extra course of prednisolone . A per protocol analysis was done on the patients who took the complete course of treatment and did not receive additional prednisolone ( 134 ) . Outcomes for skin , motor and sensory testing were compared for patients in the four treatment arms . Median improvement scores were calculated and the differences tested for statistical significance using the Kruskal-Wallis test . Survival curves using Kaplan-Meier method for recurrence events were calculated for each arm .
Three hundred and forty-five patients ( 86 . 1% males and 13 . 9% female ) , age range 15 to 60 years , were recruited from the four centres . Three hundred and thirty-six ( 97 . 4% ) of the patients had Multi Bacillary leprosy and Ridley-Jopling types Borderline Tuberculoid ( BT ) 62 . 3% , Borderline Borderline ( BB ) 5 . 5% , Borderline Lepromatous ( BL ) 21 . 7% , and polar Lepromatous ( LL ) 5 . 8% . 80% of the patients were starting MDT ( 40% ) or on MDT ( 40% ) and 20% had completed MDT . Table 1 shows that 345 patients recruited had similar distributions of skin reaction and/or the presence of sensory and motor impairments and skin reaction in each centre and for each trial arm . Skin reaction alone ( 36 . 8% ) was the commonest reason for recruitment , followed by new sensory and/or motor damage alone . Thirty-six ( 4% ) had new nerve function impairment . Two hundred patients completed the study . Fig 1 Azathioprine study Flow chart shows the outcome ( completed , withdrawn , defaulted and died ) for patients in each treatment arm . Default ( 20–26% ) and withdrawal rates ( 5 . 7% to 25 . 5% ) were lowest ( 20% ) in the prednisolone only arm . Table 2 shows that Cushingoid features ( 43% ) and infections ( 35% ) were the commonest events occurring at comparable rates across the four arms . Gastrointestinal symptoms and anaemia were more common , but not significantly , in patients treated with azathioprine , and four patients on azathioprine developed psychosis . Fifty-two out of 345 patients were withdrawn due to adverse events , pregnancy and developing ENL . The withdrawal rates between the arms ranged from 9–24% . In patients developing adverse events study drugs were withheld and the patients were monitored more closely and managed according to the protocol . Anaemia was the most common reason for withdrawal ( Table 3 ) , affecting 59 . 3% patients and a significant interaction was found between patients being on azathioprine and developing anaemia ( proportion z-test , p < 0 . 05 ) . Patients taking dapsone were also at higher risk of developing anaemia , presumably due to an interaction between these two drugs . Nausea , vomiting , gastritis , loss of weight and a general ill-health were common and nine patients were withdrawn . Four patients developed infections including multi-dermatomal herpes zoster ( 2 ) , infective gastroenteritis and hepatitis . The cause of the gastroenteritis and jaundice was not identified . Four patients became pregnant during the study and the study drugs were stopped as soon as the pregnancy was diagnosed . The pregnancies were diagnosed at six weeks . Three women chose to have their pregnancies terminated and one delivered a healthy baby . Three patients developed ENL during the study . The study drugs were stopped and the patients were managed separately for their new immune complication in the clinic . Two patients had hepatic and renal function abnormalities which resolved after per protocol management . One patient developed sub-acute intestinal obstruction associated with an undescended testis and had surgery and was withdrawn from the study . Most patients were withdrawn during the 2nd ( 22% ) and 3rd ( 20% ) months and were taking azathioprine . Patients continued to be withdrawn over the next seven study months with no withdrawals in the 11th and 12th month . Four patients died during the study and their details are summarized in Table 4 . All four patients were taking azathioprine as well as steroids and MDT . For each patient who died we tried to establish the cause of death and to establish whether azathioprine , prednisolone was the likely cause of death . Patient 1 PA1: This patient developed a high fever with loose stools . The differential diagnoses included viral fever and typhoid fever . He had been in the trial for 15 days . It is unlikely that the trial drugs caused his death . Patient 2 PA92: This patient had a cardio-vascular accident ( CVA ) ( diagnosis on death certificate ) . He developed steroid induced diabetes and had dyslipidemia . He had been in the trial for 24 weeks . His steroid treatment had been stopped at 20 weeks , four weeks before his death . We attribute his death to his underlying co-morbidity exacerbated by the steroid treatment . Patient 3 SA 102: The cause of death in this lady was not established . She developed a pancytopenia after being on trial drugs for 12 weeks . Her azathioprine and MDT were stopped . She also developed progressive anaemia , high fever , and cough and there was a possibility that she had TB . An unvalidated PCR test for M . tuberculosis DNA was positive and she had hilar lymphadenopathy on chest X-ray . She was not tested for HIV . She died in another hospital in Delhi of unidentified infection . Her death was probably related to her taking azathioprine . Patient 4 CA 01: This man had a probable myocardial infarction ( MI ) and sepsis . He was on the study drugs for 22 days . He had pre-diabetes and developed sepsis after starting medication . This could have been due to the steroid treatment or the azathioprine . The death also occurred soon after starting trials drugs so it is unlikely to be due to azathioprine . The trial outcomes were assessed by comparing the baseline with end of study clinical scores for each patient in each treatment arm . An Intention to Treat Analysis was done for 279 of the 345 patients entered into the study ( excluding 66 patients who received a second course of prednisolone ) . For patients who did not complete the study ( n = 145 ) their last assessment was taken as their endpoint ) . The combined score change and then the individual score components ( skin , sensory , motor ) are presented here . Fig 2 shows the effect of treatmentin the 4 treatment arms for the 4 parameters , combined score , skin , sensory and motor scores differences for 279 patients . Three patients were worse and 65 patients had no or little change ( 0 , 0 . 5 , and 1 ) . The median change in patients in arms 1 , 2 and 3 was 3 . 0 , and 4 . 0 for patients in arm 4 . Sixteen ( 5 . 7% ) patients improved substantially with score differences 10 and more . The baseline and endpoint score differences for patients in each arm were highly significant ( p<0 . 001 ) . However treatment with azathioprine did not improve outcomes , when the azathioprine treatment groups were compared with the steroids only treatment group , singly or combined . This is indicates that the improvement was due to the steroids and azathioprine did not enhance this improvement . Fig 2 shows the skin score changes in 180 of the 279 patients who had Type 1 reaction ( 99 patients with only NFI were excluded ) . No patients deteriorated and 38 patients had no or little change ( 0 and 1 ) . The median score changes for patients in arms 1 , 2 and 4 were 3 . 0 , and arm 3 was 4 . 0 . Fifty-seven ( 31 . 6% ) patients improved substantially , with score differences of 5 and more . The baseline and endpoint score differences for patients in arm 1 , 2 , 3 and 4 were highly significant ( p<0 . 001 ) , indicating that prednisolone treatment in these arms produced improvement and adding azathioprine did not improve skin outcomes in any arm . Fig 2 shows the score differences for patients who had sensory loss ( n = 104 ) . No patients deteriorated and 69 ( 66 . 3% ) patients had no or little change ( 0 , 0 . 5 and 1 ) . The median score change for patients in arm 1 and 2 was 0 . 5 , and in arm 3 and 4 was 1 . 0 . Seven ( 6 . 7% ) patients improved substantially with a score difference of 5 and more . The score differences were significant for patients receiving 48 weeks azathioprine ( p = 0 . 0002 ) and almost significant for patients in arms 1 and 3 ( P values 0 . 0505 and 0 . 0512 respectively ) . This suggests that azathioprine might have a beneficial effect on sensory function . Fig 2 shows the difference in motor scores for 152 patients who had motor loss . Three patients deteriorated and 72 patients had no or little change ( 0 and 1 ) . The median change for patients in arms 1 and 2 was 2 . 0 and arms 3 and 4 was 1 . 0 . Twenty ( 13 . 1% ) patients improved substantially with score differences of 5 and more . Patients treated with prednisolone only and prednisolone plus 24 weeks of azathioprine had significant improvements in their score differences p 0 . 0002 and p 0 . 0001 ) , no benefit was seen for treatment with 36 or 48 weeks of azathioprine with respect to motor function . These figures also show that a few patients in the treatment groups responded to treatment , however their numbers were not significant . A similar analysis was done on patients ( n = 253 ) who completed 20 weeks of treatment and we found similar significant improvements in the combined and skin , sensory and motor scores but azathioprine did not enhance improvement . Recurrence was calculated in the 200 patients who completed the study and 72 ( 36% ) had recurrences . There were 21 ( 32 . 8% ) , 23 ( 53 . 5% ) , 16 ( 34 . 8% ) and 12 ( 25 . 5% ) recurrences in each arm respectively . The first recurrence occurred at 16 weeks and continued throughout the study . Sixty-six patients were given a further course of prednisolone ( 18 , ( 28 . 0% ) , 21 ( 48 . 8% ) , 15 ( 32 . 6% ) and 12 ( 25 . 5% ) , and no statistical differences were found between the need for extra prednisolone between patients in the different treatment arms . Fig 3 shows the time to recurrence in each treatment arm and no significant differences were found between the four arms , showing that treatment with azathioprine did not delay the time to recurrence , although patients with the 48-week azathioprine treatment had the lowest rate of recurrence . We also did a per protocol analysis for patients who completed the treatment course ( n = 134 ) . The findings were very similar to the Intention to Treat Analysis ( ITTA ) . The combined and sensory scores are very similar in this group at 20 weeks and 48 weeks . However there was a late improvement in motor score . The ITTA ( n = 279 ) showed that there were significant baseline-end differences in scores for all the groups . Of the component scores , skin improved from baseline to end in all patients , sensory scores were only significantly improved in arm 4 and motor scores in arms 1 and 2 . None of these differences were significantly different between the steroid and the steroid + azathioprine treatment arms . So prednisolone produces the benefit and azathioprine does not add to it . The small changes in treatments arms suggest that addition of azathioprine has a small effect in these patients . Fig 4 shows the score differences for patients in each treatment arm and for the different parameters measured . The per protocol analysis show a similar picture with significant combined , skin and motor score changes from baseline to end for all treatment arms ( Fig 4 ) . Prednisolone treatment does not produce significant changes in sensory scores . This means that in this analysis for combined , skin and motor scores , prednisolone has an effect that azathioprine does not augment . For the sensory scores prednisolone is not effective . This also concurs with our finding that 67% of patients in the trial did not improve their sensory scores . The 20 weeks analysis shows a significant benefit for the combined score and a significant benefit for azathioprine treatment . The skin scores in this group all improved , the sensory scores for arms 1 , 2 and 4 and the motor improvement were all significant . The differences between azathioprine and prednisolone only treated patients were only significant for the 24 week azathioprine in the motor score . These findings can be summarised as showing that prednisolone contributes the most to improvement and there is some evidence that azathioprine gives some improvement to the motor function but azathioprine does not improve sensory function . Table 5 shows the nerve function outcomes at endpoint for all the peripheral nerve trunks by sensory and motor function . We compared the outcomes for each peripheral nerve trunk for patients receiving either prednisolone alone or APC . Each nerve was categorized into ‘unchanged’ , ‘improved’ or ‘recovered’ separately for motor and sensory score differences at the end of the study No significant differences were found with the azathioprine treatment . The ulnar , posterior tibial and median nerves improved had the most improved sensory function and the median , lateral popliteal and ulnar nerves had the most improved motor function . We hypothesized that the least affected nerve might show higher rates of recovery and the worst affected nerves the least improvement . These nerves were categorized by worst affected and least affected ( definitions in Materials and Methods ) . Fifty-seven nerves were identified in the least affected category with recovery rates of 83% of nerves in the azathioprine + prednisolone treated groups and 78% in the prednisolone treated patients . Recovery occurred in all nerves with no nerve dominating . However there were only a few nerves in each category . In the worst affected nerve group a significant benefit for azathioprine treatment was found , with 82% worst nerves treated with Azathioprine plus steroids recovered and improved whereas only 66% worst affected nerves treated with steroids recovering ( Z test p 0 . 0171 ) . We also tested the hypothesis that old nerve damage would show less improvement , by looking at the score differences for nerves that had been classified as having old damage . The sensory scores of three patients deteriorated and 72 patients had no or little change ( 0 , 0 . 5 and 1 ) . The median change for patients in all four arms was 0 . 0 . One out of 82 patients improved substantially with score difference 5 and more . Motor scores of five patients deteriorated and 39 patients had no or little change ( 0 and 1 ) . Two patients improved by 5 and more . The median change for patients in arm 1 , 3 and 4 was 0 . 0 , and arm 2 was 1 . 5 . Treatment with azathioprine did not affect old damage for either motor or sensory scores ( comparison made between arm 1 and 2 , 1 and 3 , 1 and 4 ) . This shows that most patients with old motor and/or sensory nerve damage do not improve . We analysed the score changes for all patients in the ITTA and found the following levels of change percentage wise for the combined treatment groups with the following categories , worse , static , improved and improved substantially . Combined ( 1 . 07 , 23 . 2 , 69 . 8 , 5 . 7 ) Skin ( 0 . 0 , 21 . 1 , 47 . 3 , 31 . 6 ) Sensory ( 0 , 66 . 3 , 27 . 0 , 6 . 7 ) Motor ( 1 . 97 , 47 . 3 , 37 . 6 , 13 . 1 ) Fig 5 shows the score changes for the whole cohort of the Intention to treat analysis by type of change and for each parameter combined , skin sensory and motor )
This was an important study because it is the first and largest RCT on a new treatment for leprosy T1R and nerve damage to be done with validated scores and validated tools for the skin and neurological outcome measures . [12] This trial was done to see if adding azathioprine to prednisolone could improve skin and nerve inflammation in leprosy reactions . A significant number of patients benefitted from the APC treatment with clinical score improvement but this was not significantly better than treating with P alone as analysed by both an Intention to Treat and per protocol analysis . We found that a 48 week course of azathioprine with prednisolone improves outcomes in sensory function , for motor outcomes we found that a 24 week course improved outcomes but was not superior to prednisolone alone . The effect was not seen consistently with the different lengths of treatment . The strengths and limitations of our study and the need for further studies will be discussed . Prevention of recurrence was our other outcome and 37% patients had a recurrence and so needed a further course of prednisolone . Our study design tested three different lengths of treatment with azathioprine , this enabled us to test the effects of different durations of azathioprine , but we ended up with small numbers of patients in each treatment arm and even smaller numbers in each subgroup . Our main positive findings came from the analysis done of all patients who completed the twenty weeks of treatment . The effect of azathioprine is probably small; the previous study showed that azathioprine prednisolone combinations were only slightly better than prednisolone alone . So our study was probably under-powered to detect effects due to azathioprine . We might have under-detected benefits vis a vis sensory scores because we used only one monofilament the 2gm on the hand ( 2gm ) and 10gm on the feet monofilament for sensory testing . Other studies have used the 2gm and 10gm on the hands and 10gm and 300gm on the feet . [6] The advantage of using two monofilaments or more is that one can detect a larger range of changes in sensory perception on the hands and feet . In our study we could not identify someone who could detect a 10gm but not a 2gm monofilament . We set a high threshold for sensory impairment . The poor improvement rate of 50% for motor improvements might be related to the difficulty of detecting small changes in muscle power . In the INFIR study a comparison of the ways of testing muscle power showed that manual testing was less sensitive than nerve conduction studies . [13] We used a fixed dose of 50mg azathioprine to facilitate the running of the trial . Our patient weights ranged from 32 – 95kg which translates into a weight adjusted dose of about 1mg/kg . This is a low dose but is used in patients with transplants and rheumatology conditions . Although azathioprine was associated with adverse effects , the significant and serious ones were associated with steroid treatment . We found a significantly higher rate of anaemia in patients taking both azathioprine and dapsone . This is a novel finding and will be reported in detail elsewhere . There were also four deaths , all in patients taking the prednisolone- azathiopirine combination and we were not able to identify a definitive cause of death for these patients . Two patients died within 22 days of starting azathioprine , so it is unlikely to have caused their deaths . However defining the cause of death was hampered by the absence of diagnostic and microbiological tests . Two patients may have had undiagnosed infections . Two of the hospitals ( Champa and Purulia ) are situated in remote areas without referral hospitals where complicated cases could be sent . Treatment with steroids is associated with development of new and worsening infections and diabetes . Screening for undiagnosed diabetes was part of the protocol and both patients had normal pre-trial glucose levels . So the steroid treatment might have caused the new diabetes since two patients had high peri-mortem glucose levels . The steroid part of the treatment regimen might have contributed to the cause of death in three patients , the fourth who developed pancytopenia was probably due to azathioprine . It was surprising that azathioprine did not reduce the recurrence rate and the overall rate was 37% with no difference between the treatment arms , although the 48 week azathioprine course was associated with a lower recurrence rate ( non-significant ) . In the INFIR study 30% of the study cohort developed a T1R or new nerve damage after the study started and in the methyl prednisolone study on the treatment of reactions 50% of the study participants had a recurrence . [14] [6] The recurrence rate of 36% highlights a major problem in the treatment of leprosy . [15]This phenomenon is linked to the persistent of inflammation in leprosy and strengthens the case for further studies on the immuno-pathology of leprosy and identifying the factors that contribute to continuing inflammation . This study gives important trial data about the effect of steroids in leprosy reactions . These data have been collected in the trial setting and so are more robust than data collected from observational studies . These data show that the effect of steroids on skin inflammation is significant , with skin lesion inflammation improving significantly with 78 . 9% of patients showing some improvement . However both P and the APC were relatively ineffective at improving neurological outcomes; for sensory scores 66 . 3% patients were unchanged after 48 weeks and a further 33 . 7% had a change of only 1–0 . 5 points . The motor scores showed more improvement with 47% static and 50 . 7% improved . This replicates a Cochrane review which identified only three trials on steroids and nerve damage and found no benefit from steroid treatment . [4] Our data show that new nerve damage is most likely to improve , whilst patients with old nerve damage were unlikely to respond . This confirms previous findings by the Tripod 2 and 3 trials . [16 , 17] These data are also important and novel in showing that is a significant adverse effects associated with steroid treatment . Forty five percent of patients had an adverse effect with Cushingoid features , infections and GI symptoms being common . These have not been reported in leprosy patients previously . Previous trials have not collected data on adverse effects systematically . In the methyl prednisolone study in Nepal data was collected and Cushingoid features were noted regularly . Observational studies will also miss collecting data on adverse effects and even deaths can be missed there . A recent study from Addis Ababa shows that when leprosy patients are treated with steroids rather than Thalidomide for ENL they have a mortality rate of 8% . This is further evidence on the importance of monitoring steroid use in leprosy settings carefully and being cautious . Osteoporosis is also a well-recognized adverse effect of steroid treatment . These patients were probably Vitamin D deficient and again highlight the possibility of unrecognized adverse effects . Our data also show that it is difficult to switch off inflammation in the nerves , even with steroid treatment the improvement rates were very small . It was encouraging that few patients deteriorated once on steroid treatment . Previous studies which have looked at cytokine production in skin biopsies from patients with leprosy reactions have shown that inflammation can persist longer than 20 weeks . [18] Other studies have shown that cytokines have little effect on cytokine production in skin reactional lesion . [19] These studies and our data highlight the need for studies directed at understanding the mechanisms of inflammation in leprosy so that better tools for down regulating inflammation can be developed . The strengths of this study are that the setting for treating leprosy patients was typical for patients in India , so our findings are applicable to other centres . We also had a very dedicated team who spent a lot of time following up patients and encouraging those with adverse effects . Several research needs have been highlighted by this study . Trials are needed to identify the optimum duration of steroid treatment; this might vary for skin and nerve reactions . [21] Skin reactions might be successfully treated with a shorter course of steroids such as 12 weeks long . The Tenlep study is comparing 20 versus 32 weeks of treatment with steroids and this will guide future studies . Future work also needs to focus on trying to identify patients who do not respond to steroid treatment so that their treatment can be stopped or not implemented . Molecular markers might be one useful approach . A transcriptome approach identified several novel genes in leprosy reactions . [20]This could be developed . The 35% of patients who have a relapse are also a challenge and other trials might focus on the possible benefit of giving azathioprine to these patients and determining whether they could then be given a lower dose of steroids . Other immune-suppressants have not been used systematically in the treatment of Type 1 leprosy reactions . We suggest that further small studies be done comparing prednisolone versus azathioprine plus prednisolone , given at a higher dose and for 48 weeks . More sensitive tools for measuring nerve function should be used , including a range of monofilaments for sensory testing and perhaps nerve conduction studies for assessing motor function . When using azathioprine in leprosy patients on MDT we recommend that a non dapsone containing regimen is used to minimize the development of anaemia . The service implications of our findings are that we should highlight awareness of the adverse effects of steroids when used in the field . We should also work to promote early detection of leprosy so that patients present before they have nerve damage . Future work should develop algorithms for identifying patients who do not respond to steroids and stopping their treatments rather than giving them long useless treatment courses . In conclusion we have shown that it is difficult to improve on steroid treatment for leprosy inflammation . New approaches are needed to identify the underlying mechanisms and to develop new treatments . | Type 1 reactions affect leprosy patients and are due to increased inflammation in skin and nerves that can cause disfiguring skin lesions and loss of sensation and loss of power in the hands and the feet . These disabilities can lead to deformity and severe disability . It is important to improve the treatment for T1 reactions . T1R are currently treated with steroid tablets and about 50% patients will have improvement in their nerve function after treatment . Azathioprine is a cheap widely available immune-suppressant and we tested whether it could improve skin and nerve function in leprosy patients . We did a randomised double blind study in four leprosy hospitals in India giving 345 patients treatment with steroids plus either azathioprine or placebo . 78% of patients had improved skin , 35% had improved sensory and 50% had improved motor nerve function at the end of treatment . Treatment with azathioprine did not increase patient improvement and the improvements we found were associated with steroid treatment . There was a high rate of adverse effects from both steroids and azathioprine . These findings highlight the difficulty in switching off leprosy inflammation and the need for better treatments for reactions and nerve damage . The problems of steroids causing adverse effects in patients needs to be highlighted in leprosy programmes . Research is needed to identify patients who do not respond to steroid treatment and develop alternative treatments for them . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"chemical",
"compounds",
"clinical",
"research",
"design",
"pathology",
"and",
"laboratory",
"medicine",
"immunology",
"tropical",
"diseases",
"neuroscience",
"organic",
"compounds",
"anemia",
"bacterial",
"diseases",
"research",
"design",
"cognitive",
"neuroscience",
"steroids",
"signs",
"and",
"symptoms",
"pharmaceutics",
"neglected",
"tropical",
"diseases",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"inflammation",
"lesions",
"chemistry",
"adverse",
"events",
"steroid",
"therapy",
"hematology",
"immune",
"response",
"diagnostic",
"medicine",
"organic",
"chemistry",
"leprosy",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"drug",
"therapy",
"cognitive",
"science",
"motor",
"reactions"
] | 2017 | AZALEP a randomized controlled trial of azathioprine to treat leprosy nerve damage and Type 1 reactions in India: Main findings |
Scabies afflicts millions of people worldwide , but it is very difficult to diagnose by the usual skin scrape test , and a presumptive diagnosis is often made based on clinical signs such as rash and intense itch . A sensitive and specific blood test to detect scabies would allow a physician to quickly make a correct diagnosis . Our objective was to profile the mite-specific antibodies present in the sera of patients with ordinary scabies . Sera of 91 patients were screened for Ig , IgD , IgE , IgG and IgM antibodies to S . scabiei , as well as to the house dust mites Dermatophagoides farinae , D . pteronyssinus and Euroglyphus maynei . 45% , 27% and 2 . 2% of the patients had measurable amounts of mixed Ig , IgG and IgE that recognized scabies mite antigens . However , 73 . 6% of the scabies patients had serum IgM that recognized scabies proteins , and all except two of them also had IgM that recognized all of the three species of dust mites . No patient had serum antibody exclusively reactive to scabies mite antigens . Co-sensitization or cross-reactivity between antigens from scabies and house dust mites confounds developing a blood test for scabies .
Scabies , caused by the mite , Sarcoptes scabiei , is a worldwide-occurring parasitic skin disease [1 , 2] . It was recently added to the list of neglected tropical diseases by the World Health Organization [3] . Estimates of the prevalence of scabies range from a small percent of the population in developed countries to high prevalence in some resource-limited communities in countries of the global south where the disease may affect up to 50% of women and children [4–13] . In addition , outbreaks are reported in nursing homes and daycare facilities ( among both workers and clients ) , as well as in kindergartens , hostels , schools , among colleges students , and in work environments where there is much physical contact between individuals [14–18] . Ordinary—in contrast to crusted—scabies is very difficult to diagnose . Parasitological techniques are rather insensitive and clinically scabies can mimic other skin diseases such as eczema , psoriasis , atopic dermatitis , diaper rash , poison ivy dermatitis and skin reactions to irritating agents such as soaps/detergents , metals , and lotions . A presumptive diagnosis of scabies requires confirmation by recovering mites , mite fecal pellets , and eggs from the corneal layer of the epidermis . In practice though , patients are often diagnosed with scabies based on clinical characteristics such as a rash and intense itch . Since none of the symptoms and signs are pathognomonic , this approach frequently results in a false-positive diagnosis , which in turn exposes the patient to a potentially hazardous treatment . In contrast , crusted scabies ( also known as “Norwegian scabies” ) is easily recognizable by the hyperkeratosis that manifests with a scaly and thickened ( i . e . , crusted ) strateum corneum and accompanying large mite burden as compared to the low mite burden and rash of ordinary scabies . A sensitive and specific blood test to detect scabies-specific circulating antibodies would allow a physician to quickly make a correct diagnosis . One of the problems with developing such a test is that many of the antigens from scabies mites cross-react with antigens from the common allergy-causing house dust mites , Dermatophagoides farinae , D . pteronyssinus and Euroglyphus maynei that occur worldwide [19–22] . We report here the antibody isotype profiles of the sera of two groups of patients with ordinary scabies ( from Brazil and the United States ) that recognize scabies and house dust mite antigens and illustrate the confounding problem of cross-reactivity and cross-sensitization .
Serum from the US patients ( 17 subjects + positive reference ) was collected under Human Subjects Protocol ( HSP ) #0205 as approved by the Wright State University Institutional Review Board ( IRB ) . Negative control sera were previously provided to us without personal identifiers under protocol SC #2714 approved as EXEMPT under CFR 46 . 101 ( b ) ( 4 ) by the Wright State University IRB . Serum from the Brazilian patients was provided to Wright State researchers without personal identifiers . The research was approved as EXEMPT under CFR 46 . 101 ( b ) ( 4 ) and was approved under protocol SC #4334 by the Wright State University IRB . The original study was approved by Universidade Federal do Céara—Comité de Ética em Pesquisa , protocol 358/08 . The text of the consent form was read out loud in the presence of other household members ( for both children and adults ) and the collection procedures and subsequent laboratory techniques were explained in plain language . All adult subjects and guardians of minor children then signed the consent forms that had been read to them . Sera were obtained following consent from two groups of patients with ordinary scabies at the time of initial diagnosis . Sera were collected from patients at a dermatologist’s office in Cincinnati , OH , USA . This group was composed of 17 subjects ( 6 females + 11 males , 18–72 yrs of age ) who had ordinary scabies confirmed by the recovery of live mites by skin scraping at the time of diagnosis and who reported having had symptoms for 0–13 months prior to diagnosis . Sera were also collected from 74 scabies patients ( 44 females + 30 males , 5–73 yrs of age ) in a resource-poor community in Fortaleza , Northeast Brazil . These patients were identified by active case detection . The clinical diagnosis was confirmed by dermoscopy and skin scraping and 86 . 3% of the patients had 3 or more topographic areas affected . In 42 . 1% the duration of the infestation was < 3 weeks . As a positive reference , serum previously collected from an ordinary scabies patient was used . This 72 yr old male who presented at a dermatologist’s office in Dayton , OH reported having had scabies symptoms for > 4 yrs . His serum had previously been demonstrated to contain high levels of circulating antibody to scabies [23] and radioallergosorbent testing ( RAST; conducted by Clinical Immunology and Allergy , Liberty , MO ) showed total IgE > 1000 U/mL and modified RAST class 2 scores ( scores range from 0 to 6 ) for specific IgE to both D . farinae and D . pteronyssinus . A pool of negative control sera was prepared by mixing equal volumes of serum from two individuals that had no known history of scabies and that had previously been demonstrated to have low levels of circulating antibody to extracts of any of 9 astigmatid mite species [24 , 25] . Aqueous extracts were prepared from Sarcoptes scabiei var . canis and from the house dust mites Dermatophagoides farinae , D . pteronyssinus , and Euroglyphus maynei according to our standard protocol . Briefly , scabies mites were collected by aspiration onto a 38 μm stainless steel mesh ( Small parts , Inc . , Miami Lakes , FL ) after they had migrated from crusts . To remove host material , live mites were washed by drawing sequential 4 mL aliquots of PBST ( Dulbecco’s Phosphate Buffered saline with 0 . 05% Tween 20 ) , endotoxin-free water and 70% ethanol through the mesh . Mites were killed by freezing at -80°C where they were stored until used . Dust mites were collected by aspiration as they migrated from cultures , killed by freezing , lyophilized and stored at -20°C until used . Mites of each species were suspended in endotoxin-free water ( at 1:10 W:V for dust mites and 1:20 W:V for scabies mites ) for overnight extraction at 4°C . The next day , samples were ground 10 strokes on ice using a TenBroeck homogenizer , the insoluble material was removed by centrifugation for 10 min at 14 k x g and the supernatants were sterile filtered into sterile vials . To extract as much protein as possible , the scabies mite pellet was subjected to a second extraction/homogenization/centrifugation and the two supernatants were combined . Protein content in each extract was determined by the method of Bradford [26] using bovine serum albumin ( BSA ) as the standard . Microtiter plates for each mite extract were coated with 0 . 5 μg protein/well in 50 mM carbonate/bicarbonate buffer , pH 9 . 6 , and allowed to dry . Plates were made in two batches ( one for each serum cohort ) and were stored in a desiccator at room temperature until used . Before use and between all steps , plates were washed with PBST . Plates were blocked with 1% BSA in PBST and loaded with 100 μL of serum diluted 1/10 ( IgD and IgE ) or 1/1000 ( Ig , IgG and IgM ) . All samples were loaded in duplicate wells and each plate had a parallel set of positive a4nd negative control sera . Serum antibody binding was detected using biotinylated antibodies specific for human mixed Ig [IgM+IgG+IgA , H+L] , IgD [δ chain specific] , IgE [specific for Fc portion of the heavy chain] , IgG [γ chain specific] , or IgM [μ chain specific] ( all diluted 1/5000 ) followed by streptavidin-peroxidase ( 1/5000 ) . All were purchased from Southern Biotech ( Birmingham , AL ) . Plates were developed with 100 μL of 1 mM ABTS in 70 mM citrate phosphate buffer , pH 4 . 0 , with 0 . 03% hydrogen peroxide . Development was stopped by the addition of 50 μL of 0 . 2% sodium azide and plates were read immediately . To account for plate-to-plate variability , all data were normalized at the conclusion of the study . For each antibody class and antigen , the 405 nm absorbances of all the wells for the positive control serum were averaged . These values were then used , along with the controls for each individual plate , to calculate normalized ELISA absorbance values for each serum/antibody class/antigen .
Among the 17 US patients , 10 had IgM , 6 had IgG and 6 had Ig that recognized scabies mite antigens . Five patients with elevated IgM antibodies to scabies antigens did not have elevated IgG or mixed Ig to scabies . Conversely , 3 patients had elevated IgG antibodies to scabies but not elevated IgM and 3 patients did not show any antibody binding to scabies mite proteins . No US patient had detectable levels of IgE to scabies mites . While not all of the US patients had circulating antibody that bound to scabies antigens , all did have elevated levels of at least one antibody class that recognized antigens of the house dust mites D . farinae , D . pteronyssinus , and E . maynei . All 6 patients with Ig to S . scabiei also had Ig to D . farinae , D . pteronyssinus , and E . maynei . There was also no IgE binding detected to any of the house dust mite extracts . Among the 74 Brazilian patients , 57 had IgM and 19 had IgG that recognized scabies antigens while 35 patients had elevated mixed Ig to scabies . Of the 57 patients that had elevated IgM to scabies , 23 did not have elevated IgG or mixed Ig to scabies antigens . Two of the Brazilian patients had detectable levels of IgE to scabies antigens . Eight of the scabies patients had no detectable circulating antibody binding to scabies antigens . With the exception of one patient that showed no antibody binding to D . pteronyssinus , all Brazilian patients had Ig , IgG and/or IgM that bound to antigens of all three house dust mite species . Of the 57 patients that had elevated IgM to scabies all except one also had elevated IgM to D . farinae , D . pteronyssinus and E . maynei . Likewise , of the 19 patients that had elevated IgG to scabies all also had elevated IgG to all three dust mites . Of the 35 patients with mixed Ig that recognized scabies mite antigen , all but two also had Ig that recognized antigen of all three species of dust mites .
Although scabies mites burrow and reside in the stratum corneum of the epidermis , historical studies along with more recent studies clearly indicate that antigens from scabies mites induce a humoral response in the host . Thus , theoretically it may be possible to develop a blood test for scabies based on circulating antibodies that recognize scabies-specific antigens that do not cross-react with house dust mites . We provide a brief history of studies profiling information that leads to the conclusion that scabies can be diagnosed with a simple blood test but that the antigens will need to be carefully selected . Some early studies reported highly diverging results concerning total serum IgG , IgA , and IgE , and complement C3 and C4 levels in scabies patients compared to control subjects without scabies [27–31] . None of these studies directly investigated whether or not the altered serum immunoglobulin isotype levels were the result of scabies infestation and were specific to scabies mite antigens . Hence , these studies do not allow any conclusion . Several studies indirectly associated serum antibody concentrations to scabies infestation . These studies showing a relationship between changes in antibody isotype levels during scabies infestation compared to those following successful treatment suggested a cause-and-effect association with scabies infestation although the presence of specific antibodies to scabies antigens were not determined [28 , 32 , 33] . However , it was not excluded that the scabies patients were concomitantly infected with helminths and that the presence of helminths caused the immunological alterations . More recent studies using S . scabiei mite extract or recombinant scabies molecules prepared from mites collected from different host species clearly showed that animal and human hosts build antibodies to specific S . scabiei molecules [21 , 23 , 34–42] . Two studies showed that human patients with crusted scabies had elevated total scabies-specific IgE compared to those with ordinary scabies [35 , 43] . Walton et al . [43] found that subjects with both ordinary and crusted scabies had elevated antibody levels specific for several recombinant scabies antigens compared to naïve control subjects never exposed to scabies . Additionally , an immunoassay employing a recombinant scabies protein , rSar s 14 . 3 , that corresponds to residues 1263–1655 of the dust mite allergen Der p 14 , showed significantly higher IgE binding by plasma from crusted scabies patients than from ordinary scabies patients [44] . The IgE binding by the ordinary scabies patients was also significantly higher than that of controls and no cross-reactivity with the dust mite homolog was observed . Obtaining sufficient amounts of human scabies mites for research purposes is very difficult . However , use of mites collected from dogs , pigs , and foxes as a source of material for research to develop a diagnostic test and vaccine for human scabies offers a promising alternative . It is not clear if S . scabiei mites from such hosts are the same species or only subtle genetic variants because S . scabiei mites from most different host species are morphologically indistinguishable . The most recent molecular study indicated that mites collected from humans could be distinguished from those collected from animals based on sequencing of the 317-bp mtDNA cox1 gene but not when several other molecular markers were used [45] . This analysis also suggested that all the mites collected from various animal hosts were monospecific while the mites isolated from humans from different geographical locations clustered into four separate branches representing four different species [45] . More importantly , several studies have shown that there is significant antigenic cross-reactivity between different strains of scabies mites from different hosts but there are also strain-specific antigens [23 , 46] . For example , S . scabiei var . suis from pigs , var . canis from dogs and var . hominis from humans share cross-reacting antigens . Likewise , humans with scabies have circulating antibodies that recognize antigens from S . scabiei var . vulpis from foxes , var . canis , and var . suis [23 , 46] . Haas et al . [46] found that 48% of scabies patients had IgG that recognized antigens from fox mites and patients with greater severity and duration of scabies had significantly higher IgG titers . In addition , Western blotting showed that a 72-year-old male with chronic scabies had IgG that recognized more than 10 antigens from S . scabiei var . canis [23] . Also , serum from 6 patients with crusted scabies had IgE to 11–21 and IgG to 1–7 antigens from S . scabiei var . canis but patients with ordinary scabies had serum IgG and IgE that recognized many fewer antigens [35] . And a crude whole body extract of S . scabiei var . vulpes obtained from red foxes contained antigens recognized by antibodies in serum from scabies infested pigs [47] and chamois [48] . An ELISA using the recombinant Ssλ20ΔB3 antigen from S . scabiei var . hominis detects serum antibody in Iberian red deer , Southern chamois , pigs , and rabbits infected with sarcoptic mange [49–53] . Likewise , several recombinant proteins generated from S . scabiei var . suis were recognized by antibodies in sera of human patients infected with S . scabiei var . hominis [54] . Therefore , using mites collected from various host species provides a means of identifying specific molecules and antibody isotypes that may be useful in developing a diagnostic test for scabies and overcomes the problem of a limited supply of var . hominis mites . The present study using a canine strain of scabies to prepare antigen for use in ELISA , found that only 45% , 27% and 2 . 2% of 91 patients with ordinary scabies had measurable amounts of mixed Ig , IgG and IgE that recognized scabies mite antigens , respectively . No patient had IgD that recognized scabies mite antigen . However , 73 . 6% of the scabies patients had serum IgM that recognized S . scabiei antigens with substantially higher levels observed from the Brazilian patients compared to those from the US . This suggests that the Brazilian patients were diagnosed rather early in the infestation since they had not switched to IgG production that follows an initial IgM response . Conversely , the observation that more patients had IgG to the house dust mites suggests a more chronic exposure to these mites . Based on these results , it appears that a diagnostic test should be based on detecting serum IgM to scabies antigens . Such a test would be beneficial because IgM is the first antibody class produced , before class switching to IgG occurs , so it may allow for earlier diagnosis of scabies . The differences between the responses of the two groups of patients may reflect differences in the strains/species of scabies mites infesting these geographically-distant patients as suggested by the Zhao et al . study [45] . This study again elucidates the problem of co-sensitization or cross-reactivity between antigens from house dust mites that confounds developing a blood test for scabies . House dust mites are the sources of > 25 antigenic proteins [55] . Many of the antigens from house dust mites cross-react with those from scabies mites [19–22] . Also a significant percentage of people are sensitized to the ubiquitous house dust mites , D . farinae , D . pteronyssinus and E . maynei . Every scabies patient of this study except one had circulating Ig , IgG , and/or IgM to all three house dust mite extracts and no scabies patient had antibodies exclusively to scabies mites . Thus , the key to a diagnostic test that possesses both specificity and sensitivity likely lies in identifying a limited and defined set ( used as a cocktail ) of scabies proteins ( or protein fragments ) that do not carry epitopes that cross-react with those on house dust mite proteins and that bind IgM from the serum of scabies patients . To accomplish this goal , a detailed and comprehensive proteomic and genomic analysis of S . scabiei is necessary . | Scabies , caused by the mite S . scabiei that burrows in the skin of humans , is a contagious skin disease that affects millions of people worldwide . It is a significant public health burden in economically disadvantaged populations , and outbreaks are common in nursing homes , daycare facilities , schools and workplaces in developed countries . It causes significant morbidity , and in chronic cases , associated bacterial infections can lead to renal and cardiac diseases . Scabies is very difficult to diagnose by the usual skin scrape test , and a presumptive diagnosis is often made based on clinical signs such as rash and itch that can mimic other skin disease . A sensitive and specific blood test to detect scabies-specific antibodies would allow a physician to quickly make a correct diagnosis . Our manuscript reports the antibody isotype profiles of the sera of two groups of patients with ordinary scabies ( 17 from the US and 74 from Brazil ) and concludes that such a blood test should be based on circulating IgM type antibodies that do not also recognize antigens of the related and ubiquitous house dust mites . Both are important considerations for research for developing a blood test for the diagnosis of scabies . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | The Potential for a Blood Test for Scabies |
The transition from single-cell to multicellular behavior is important in early development but rarely studied . The starvation-induced aggregation of the social amoeba Dictyostelium discoideum into a multicellular slug is known to result from single-cell chemotaxis towards emitted pulses of cyclic adenosine monophosphate ( cAMP ) . However , how exactly do transient , short-range chemical gradients lead to coherent collective movement at a macroscopic scale ? Here , we developed a multiscale model verified by quantitative microscopy to describe behaviors ranging widely from chemotaxis and excitability of individual cells to aggregation of thousands of cells . To better understand the mechanism of long-range cell—cell communication and hence aggregation , we analyzed cell—cell correlations , showing evidence of self-organization at the onset of aggregation ( as opposed to following a leader cell ) . Surprisingly , cell collectives , despite their finite size , show features of criticality known from phase transitions in physical systems . By comparing wild-type and mutant cells with impaired aggregation , we found the longest cell—cell communication distance in wild-type cells , suggesting that criticality provides an adaptive advantage and optimally sized aggregates for the dispersal of spores .
Many living systems exhibit collective behavior , leading to beautiful patterns found in nature . Collective behavior is most obvious in animal groups with clear advantages in terms of mating , protection , foraging , and other decision-making processes [1 , 2] . However , how cells form collectives without visual cues is less well understood [3] . There are two main strategies to achieve synchrony ( or long-range order ) among individuals: A leader ( i . e . , a special cell or an external chemical field ) may influence the behavior of the others in a hierarchical fashion ( top-down ) . An example is the developing fruit fly embryo in a maternally provided morphogen gradient [4 , 5] . Alternatively , all individuals are equivalent and order emerges spontaneously by self-organization ( bottom-up ) . Examples may include organoids [6] and other cell clusters [7] , and both strategies are not mutually exclusive . While order itself cannot be used to differentiate between the two mechanisms , the response to perturbations or , simply , the correlations among fluctuations can be examined [8] . In top-down ordering , fluctuations are independent as cells follow the leader or the external field , and hence they are not influenced by their neighbors . In contrast , in bottom-up ordering , cells are coupled to their neighbors . Hence , fluctuations are correlated as neighboring cells influence each other [9] . Note that in this context , it is a reasonable assumption that cells can follow fluctuations of their neighbors much more easily than fluctuations of a distant leader cell . At a critical value of the cell—cell coupling strength , correlations may establish among cells that span the whole cell collective independent of its size , leading to a maximally connected collective similar to neurons in the brain [10] . To test these ideas of achieving order and long-range communication , we considered the well-known social amoeba Dictyostelium discoideum , which undergoes aggregation in response to starvation [11–13] . During this developmental process , cells start to secrete pulses of cyclic adenosine monophosphate ( cAMP ) , a molecule that also acts as a chemoattractant for the other cells in the vicinity . The underlying signaling and regulatory pathways of such development have been thoroughly examined using genetics and imaging [14]: when a cell is “hit” by a high concentration of cAMP , it secretes a pulse of cAMP itself , relaying the signal and thus causing the formation of cAMP waves , which are inferred indirectly from optical density waves in dark-field movies [15 , 16] . These waves propagate through the whole population [17–20] . As their development proceeds , cells pulse at higher frequencies , reaching frequencies of up to one pulse every 5 min in the aggregate [21 , 22] . Cell movement also accompanies the secretion process: before cells start to secrete cAMP , they normally move incoherently; when cAMP waves form , cells move towards the direction of the incoming wave by following the cells emitting the pulse in an orderly fashion ( streaming phase ) . Interestingly , in a microfluidic device , cells did not follow an artificially produced cAMP wave once it passed the cells , despite it producing a gradient behind the cells pointing in the opposite direction of cell movement . Hence , cells are thought to solve the so-called “back-of-the-wave” problem for directed unidirectional chemotaxis towards the aggregate [23 , 24] . While single-cell chemotaxis [23–28] and large-scale pattern formation [16 , 29–33] have been extensively studied , a precise characterization of the transition from single cells to multicellularity is still missing . Here , we developed a multiscale model to capture the mechanism of aggregation , focusing on the distinction between induced and self-organized order . Specifically , we were able to unify single-cell behavior and multicellularity at wide-ranging spatiotemporal scales . We achieved this by extending a single-cell model , which is able to describe Dictyostelium cell shape and behavior [25] , by adding intracellular cAMP dynamics , secretion , and extracellular dynamics for cell—cell communication . To simulate hundreds of cells , we extracted a set of minimal rules for building a coarse-grained model . Hence , our approach is able to capture all stages of aggregation , ranging from single-cell chemotaxis to the multicellular collective . For quantifying the transition from disorder ( preaggregate ) to order ( aggregate ) , we employed the mathematical concepts of spatial information and directional correlations . We found that the transition occurs during the streaming phase , which resembles a critical-like point known from phase transitions in physical systems as extracted by finite-size scaling . In physical systems , phase transitions are characterized by an abrupt change in the macroscopic properties of the system when an external parameter ( such as temperature ) crosses a well-defined value . In our cell system , this parameter is the cell density ( or external cAMP concentration ) . Criticality was tested by corresponding analyses of previously recorded time-lapse movies from fluorescence microscopy ( provided by the Gregor lab [22 , 34 , 35] ) . Comparison of different Dictyostelium strains showed that wild-type cells have a longer cell—cell communication range than any mutant strain with impaired aggregation ( based on regA and rdeA mutant data from the Cox lab [21] ) , even if cells with enhanced cell—cell adhesion ( such as cells that secrete less cell number “counting factor” [36] ) form larger clusters . Hence , criticality may give cells an adaptive advantage , leading to optimally sized aggregates .
To model the transition from single cells to multicellularity , we started with cell shape and behavior in single cells . Specifically , we considered a model capturing single-cell membrane dynamics similar to the Meinhardt model [25 , 26 , 37] . Although not based on specific molecular species , this model describes membrane protrusions ( such as pseudopods ) and retractions as well as resulting cell movement by means of three effective equations ( see Supporting information ) . The first and second variables are a local activator and a global inhibitor ( both are also considered in the local excitation , global inhibition [LEGI] model [27 , 28] ) . The third is a local inhibitor , which is important in order to destabilize the current pseudopod and to increase the responsiveness of the cell ( Fig 1A , left ) . To this we added equations representing the internal cAMP dynamics based on the FitzHugh—Nagumo model ( Fig 1A , middle ) [34 , 35] . These are meant to capture the intracellular cAMP dynamics that are governed by the relative activities of adenylyl cyclase ( ACA ) , which synthesizes cAMP , and 3ʹ , 5ʹ-cyclic-nucleotide phosphodiesterase regA , which degrades cAMP . Based on experimental evidence , we assumed that cAMP is released from the rear of the cell [38 , 39] . We also modeled extracellular cAMP dynamics for cell—cell communication , taking into account diffusion of cAMP in the extracellular medium and its degradation by secreted phosphodiesterase PDE ( Fig 1A , right; see Materials and methods for further information and Supporting information for numerical implementation ) . Using this detailed model , we investigated the behavior resulting from cell—cell interactions in very small systems . First , we wanted our model to capture streaming ( i . e . , the ability of a cell to precisely follow the cell in front of it ) . To reproduce that , we simulated a single cell in a rectangular box with periodic boundary conditions ( see Fig 1B and S1 Movie ) . In mathematics , periodic boundary conditions mean that the box is neighbored by identical copies of the box . Thus , in practice , if a cell passes through one side of the box , it reappears on the opposite side with the same velocity . This effect also applies to the molecules surrounding the cell . Now , given the rectangular shape of the box with the long side in the vertical direction and the short side in the horizontal direction , a horizontally moving cell can sense its own secretion ( as the box is neighbored by identical copies of the box and hence copies of the cell and cAMP ) . Hence , the front of the cell is able to sense the secreted cAMP at the rear of the neighboring cell . In contrast , a vertically moving cell is too far away from its rear and thus cannot sense its secretion . We estimated the ability of the cell to stream by measuring the chemotactic index ( CI ) in the x direction , calculated as the amount of movement in the horizontal direction compared to the total length of the trajectory . In Fig 1B , we show that the CI in the x direction is significantly higher than the CI in the y direction . We then considered the wave response as measured in microfluidic experiments , in which cells are exposed to traveling waves of cAMP [23 , 24] . When hit by a traveling wave , cells moved towards the direction of the incoming wave but did not follow the wave after it passed . In order to capture this robust chemotaxis behavior , our model cell undergoes a refractory period ( as was done in previous models [29 , 31] ) during which the cell can neither repolarize nor pulse ( see Supporting information for further details ) . Experimental evidence for this refractory period stems from the several-minute-long directional bias in cell polarization [42]—which may be caused by the large-scale inert cortical structure or phosphoinositide 3-kinase ( PI3K ) —that stays on the membrane even when no longer active [43] . In our simulations , this refractory state is naturally achieved when the cell spontaneously emits a pulse of cAMP upon encountering the wave ( see Fig 1C and S2 Movie ) . As a result , the CI is significantly higher in the right direction of the incoming wave . Finally , we considered a small number ( four ) of cells in a small box ( with periodic boundary conditions ) and tested whether they showed signs of aggregation ( see Fig 1D and S3 Movie ) . Specifically , we measured the tendency of the cells to cluster by calculating the density pair correlation function ( see Materials and methods ) and compared the cases with and without secretion of cAMP . In the absence of secretion , cells were randomly distributed in space at the end of the simulations , as evident by the relatively flat horizontal line of unit correlation for distances larger than one cell length ( the reduction at close distance is due to volume exclusion ) . With secretion , cells tended to be much closer to each other , with a clear peak in the density distribution at cell—cell contact ( distance of two cell radii ) , indicating that cells tend to be close to each other and hence to cluster . In order to reproduce aggregation as observed in experiments ( e . g . , [22 , 34] ) , we needed to simulate hundreds to thousands of cells . However , the detailed model introduced in the previous section is computationally too expensive , forcing us to introduce several simplifications . In our coarse-grained simulations , cells are point-like objects moving in continuous space . In particular , we took advantage of the spatiotemporal cAMP profiles from the detailed model by extracting the concentrations typically secreted by a single cell during leakage or a pulse . Shaped by degradation and diffusion , these profiles are approximately short-ranged exponential with a decaying amplitude in time . To capture the effects of volume exclusion , we also reduced the gradients in the cell-forward direction ( see Materials and methods for further information ) . Hence , as in the detailed model , the maximum cAMP concentration secreted by an individual cell is always found in the direction opposite to the direction of its motion . Using these analytical cAMP profiles , the cAMP concentration a cell senses is given by the sum of secretions by its neighboring cells . We then set concentration and gradient thresholds to determine whether a cell leaks or pulses cAMP , followed by a refractory period , and whether a cell moves randomly or follows the local cAMP gradient ( see Fig 1E , Materials and methods , and Supporting information for a detailed explanation of the model ) . Using this minimal set of rules , we simulated thousands of cells with a density similar to the experimental ones ( around a monolayer [ML] with 1 ML = 6 , 600 cells/mm2 [22 , 34] ) . Cells were initially distributed uniformly in space and allowed to move randomly . As soon as the cell density ( and hence local cAMP concentration ) increased spontaneously because of random cell motion , a cell may sense a concentration of cAMP large enough to pulse , and this excitation will propagate throughout the whole population . Because of cell movement , streaming and aggregation into a small number of clusters can be observed ( Fig 1F and S4 Movie ) . To quantify aggregation in a mathematical way , we estimated the “degree of order” ( or spatial information ) in an image . This spatial information is based on the calculation of the 2-D Shannon entropy , which does not require tracking of individual cells ( see Materials and methods for mathematical details and Supporting information for a primer on information theory ) [44] . In this framework , evenly distributed cells correspond to a low spatial information , while highly clustered cells have a high spatial information . In all simulations , we observed that the spatial information rises sharply during the streaming phase as expected for cells in an ordered aggregate ( see Fig 1F ) . Interestingly , the spatial information was previously used to capture the second-order ( disorder—order ) phase transition in the 2-D Ising model ( magnetic spins on a lattice ) [44] . Hence , we wondered whether aggregation may be viewed as a critical-like point that describes the sudden transition from individual cells to the cell collective ? Based on our model assumptions , all cells are treated the same . However , aggregation may still be driven by the first random cell pulsing ( hierarchical system ) or can spontaneously emerge as cells are coupled to each other by cAMP sensing and secretion ( self-organized system; Fig 2A ) . The order of the collective process can be measured by studying the directional correlations of pairs of cells . Specifically , the nonconnected ( nc ) correlations Cnc ( r ) =∑i≠jNui→⋅uj→δ ( r−rij ) ∑i≠jNδ ( r−rij ) ( 1 ) represent the average similarity of the direction of motion for every pair of cells depending on their distance , where N is the total number of cells , u→i is the vector representing the direction of cell i , and δ ( r − rij ) is equal to 1 if r = rij and 0 otherwise . Cnc ( r ) also represents the order parameter in our system ( i . e . , the quantity describing the degree of order or polarization in the system ) . For instance , when cells move independently of each other in random directions , then the order parameter is zero . In contrast , when all cells move in the same direction , then the order parameter would be maximal ( i . e . , one ) . By calculating this quantity for every time frame , we were able to analyze its variation in time . During the preaggregation stage , correlations are close to zero even at short distances , while they increase sharply during the streaming phase ( Fig 2B , top ) . Although order increases during the streaming phase , the origin and characteristics of this order are yet to be determined . To achieve this , we need to know more than the fact that the directions of cell movement are correlated ( which describes the degree of order even in a hierarchical , top-down system ) . In addition , we also need to know if the fluctuations of the directions are correlated . This would describe to what level cells communicate with each other and how they would respond collectively to perturbations . For this purpose , we calculated the connected ( c ) directional correlations Cc ( r ) , measuring the similarity of the directional fluctuations with respect to the average velocity [8 , 9] . For instance , Cc = 0 ( Cc = 1 ) means that a change in a cell’s direction is independent of ( perfectly matched by ) changes in the direction of its surrounding cells . To obtain the connected correlations , direction u→i in Eq 1 is substituted by the velocity of the single cell when the average is subtracted: i . e . , δui→=δvi→/1N∑j=1Nδv→j2 with δvi→=vi→−1N∑j=1Nvj→ . For this kind of collective movement , such a subtraction is not straightforward . If we compute a global average velocity for every time frame , we systematically overestimate the nonconnected correlations because we still consider part of the “bulk" velocity vectors as a result of the position of the cells in the image ( see Supporting information for a schematic explanation ) . To reduce this artefact , we considered local averages . For every cell , we considered the average velocity of all cells in its neighborhood up to a certain maximal distance rc , computed the correlations between the cell in the center and all the cells belonging to its neighborhood , and repeated this procedure for every cell in our image . When applied to the simulations , Fig 2B , bottom shows significant connected correlations , especially during streaming . Next , we considered the correlation length ξ0 ( i . e . , a cell’s “influence radius” over its surrounding cells ) . We estimated this correlation length by the minimum distance at which the correlations cross zero , i . e . , C ( r = ξ0 ) = 0 [8] . We found that ξ0 is indeed much larger than the minimum nearest-neighbor distance . This indicates that a cell influences other cells way beyond its immediate neighbors , strongly suggesting self-organization ( Fig 2C ) . Above , we demonstrated that aggregation in Dictyostelium is highly ordered and self-organized , with a correlation length much greater than the nearest-neighbor distance . Does the transition from disorder to order in this finite system show signs of criticality , given by a drastic and sudden qualitative change in behavior ? At criticality , all cells would remarkably influence each other independent of the distance between them . In order to answer this question , we considered that in critical systems , the correlation length should scale with the size of the system as there is no intrinsic length scale [8] . To investigate this , we analyzed how the correlation length ξ0 changes in time . In all simulations , ξ0 was small before aggregation and increased markedly during the streaming phase ( Fig 2C ) . In equilibrium phase transitions , the susceptibility describes how sensitive the system is to perturbations , and this quantity would diverge at the critical point for an infinitely large system . Thus , this divergence indicates that the whole ( infinite ) system responds coherently as a single unit . In our cell system , the susceptibility can approximately be computed by the integrated correlations ( i . e . , by the amount of correlated cells ) , χ=1N∑i≠jNδui→⋅δuj→θ ( ξ0−rij ) , ( 2 ) where θ ( ξ0 − rij ) is equal to 1 for rij < ξ0 and 0 otherwise [8] . This proxy for the susceptibility peaks precisely during the streaming phase ( Fig 2D , inset ) , and the higher the number of cells , the higher the susceptibility . Moreover , if we consider cell density as a control parameter ( similar to temperature or coupling in a ferromagnetic Ising model ) , we can plot χ with respect to the rescaled nearest-neighbor distance ( Fig 2D and Materials and methods ) . The resulting peak heights do not only reflect the number of cells , but their positions also shift to smaller nearest-neighbor distances ( i . e . , cells become more densely packed ) as the number of cells increases , further supporting the resemblance to a scale-free system near criticality [8] . In theory , this peak height should keep increasing with cell number and ultimately diverge at a critical nearest-neighbor distance for an infinite system . Furthermore , by normalizing the correlations ( so that they are one at the start for small distances ) and rescaling the distance by the correlation length , the correlations collapse for all our simulations ( Fig 2E ) . This collapse of the curves shows that they all have the same shape upon rescaling , indicating self-similarity as often occurs at criticality [45] . Finally , we took advantage of our image partition with different radii rc to examine how the correlation length ξ0 scales with system size . We noticed that for all movies , higher cell numbers display longer correlation lengths for a given neighborhood radius and that the correlation length increases with increasing radius . Hence , the correlation length scales with system size ( Fig 2F ) , indicating critical-like behavior in our simulated cells . To test the model , we analyzed five previously recorded movies of Dictyostelium aggregation with different cell densities from [22 , 34] ( see Materials and methods and S5 Movie ) . Briefly , during 15 h of observation , individual cells become a single , multicellular organism , going through different stages including preaggregation , streaming , and aggregation ( see Fig 3A ) . Cell densities ranged from 1/3 ML to almost 1 ML , ensuring aggregation while restricting our system to 2-D . A 10% subpopulation of cells expressing the mRFPmars ( also known as TRED ) fluorescent marker were tracked using a custom-written software ( see Materials and methods ) . Based on these cells , we repeated the analysis from the simulated cells for the TRED cells from the experiments , applying spatial information , nonconnected and connected correlations , correlation length , and susceptibility . As we used the same computational protocol for both simulations and data , a close comparison was possible , which allowed us to assess finite-size scaling and hence critical-like behavior . Based on our analysis of the data , the correlation length ξ0 increases during the streaming phase , as does the susceptibility χ ( Fig 3B–3E ) . Additionally , χ increases with cell number ( and hence cell density ) , and the nearest-neighbor distance decreases , similar to the simulations . The correlation profiles , normalized and rescaled by the nearest-neighbor distance , largely superimpose for the different cell numbers , indicating that the slope of the resulting curves is not affected by the number of cells ( see Fig 3F ) . Note that the cell density changes slightly over the duration of observation because of open boundary conditions ( the observation field is smaller than the field of cells so cells can freely move in and out of the observation field; see Supporting information for a quantification ) . Hence , cell numbers reported refer to the streaming phase ( Fig 3D , inset ) . Finally , we studied how the correlation length changes for different system sizes by considering different neighborhood radii as performed for the simulations ( see Materials and methods ) . We noticed that ξ0 increases for a given radius with increasing cell numbers and also for a fixed number of cells with increasing neighborhood radius ( Fig 3G ) . These observations strongly suggest that there is no intrinsic correlation length but that this length scales with system size . Taken together , our results suggest that aggregation can be viewed as a critical-like point in this finite system . While criticality leads to long-range cell—cell communication , what is its biological function in aggregation and ultimate spore dispersal ? This question can also be raised from the perspective of modeling: many published models achieve aggregation [29 , 31 , 46] ( although assessing potential differences in the quality of aggregation is difficult in retrospect ) . If aggregation is readily achievable , what does criticality add to aggregation ? There might be two ways to interpret this paradox: One is that all successful models are fine-tuned to achieve aggregation , and this special point is again our critical-like point . Alternatively , simple aggregation is easy to achieve , but aggregation of thousands of cells into a single aggregate ( or very few aggregates ) is difficult and requires a diverging correlation length and hence exceptionally good long-range cell—cell communication . In this context , criticality may help to make this process robust to variability and obstacles in nature , as often microscopic details do not matter near a critical point [47–49] . To address this important question , we altered model parameters or reduced earlier assumptions . Specifically , we conducted aggregation simulations of 500 cells with ( 1 ) uniform ( radially symmetric ) secretion of cAMP ( instead of secretion from the cell rear ) ; ( 2 ) increased sensing noise ( to address the naturally occurring cell-to-cell variability ) ; ( 3 ) additional cell—cell adhesion ( by the TgrA/C adhesion system during late aggregation [50] ) , which was not part of our original model; and ( 4 ) mutant cells with asynchronized cAMP secretion ( similar to the regA mutant [22] ) . Experimentally , it was found that regA and rdeA with a diminished phosphorelay ability as well as PDE mutants are able to aggregate , albeit into smaller clusters without streaming [51–55] . Furthermore , mutants with decreased “counting factor” secretion ( countin , cf45-1 , cf50 , or cf60 ) and hence increased cell—cell adhesion form larger cell clumps [36 , 54 , 56] . These cell types are implemented in our simulations as described in the Supporting information . To quantify the range of cell—cell communication and the quality of aggregation , we considered the correlation length during the streaming stage and the spatial information of the final aggregate , respectively . Fig 4 , panels A and C show that these additional modified simulations exhibit decreased correlation lengths and spatial information as compared to our previous simulations ( wild-type cells ) . Surprisingly , this even applies to the simulations with enhanced cell—cell adhesion , which produce broader aggregates as compared to wild-type cells . Thus , strong cell—cell adhesion leads to strong order , but apparently this does not allow for sufficient flexibility during the aggregation process . These findings are not in contradiction to earlier modeling , in which uniform secretion and adhesion allowed streaming to occur [57]; our results simply show that secretion from the cell rear further improves long-range cell communication and cell streaming and that early adhesion can be detrimental to streaming . Subsequently , we estimated the correlation length and spatial information from previously published movies of wild-type cells as well as regA and rdeA mutants [21 , 55] ( for details , see Supporting information ) . Similar to our simulations , we found that also in experiments wild-type cells have a larger correlation length and spatial information than any of the analyzed mutants ( Fig 4B ) . Note that the simulations with “sensing noise” and “asynchronous secretion” can be made even noisier , which would reduce the correlation length and spatial information even further to match the data better . Hence , criticality may allow wild-type cells to create aggregates of just the right size and , we speculate , may also be useful for decision making on where to aggregate as well as for increased robustness in the presence of obstacles and cell-to-cell variability . For instance , if aggregates are too small , stalks may be too short to disperse spores efficiently or too thin to support the weight of the spores [54] . In contrast , if aggregates are too large , then cells may encounter difficulties in decision making or cell sorting , or their stalks may collapse under the weight of too many spores . Hence , criticality may allow cells to make informed decisions for achieving optimal aggregate sizes for the most effective spore dispersal . This would indicate that criticality constitutes an adaptive advantage .
Dictyostelium aggregation represents a fascinating example of synchronous collective cell behavior , spanning ∼1 mm in length although cells are just ∼10 μm in size . Here , we asked how cells achieve such exquisite long-range communication [58] , when the transition from single cells to the a collective occurs , and how this transition can be characterized quantitatively . To capture the main features of aggregation , we developed a multiscale model . First , we focused on single cells using a detailed model combining sensing , cell-shape changes , and movement with cAMP secretion or pulsing and hence cell—cell communication . Once this model resembled the behavior of a single cell or a small group of cells , it allowed us to extract a minimal set of rules that could lead to aggregation . In particular , we extracted the cAMP concentration profile of a pulse from the detailed simulations and the refractory period after pulsing . By allowing cells to leak cAMP and to randomly move below a certain cAMP threshold concentration , we were able to observe spontaneous random pulsing as soon as the local density increased , similar to what occurs in real cells . This minimal set was subsequently included in the coarse-grained agent-based model , which is able to reproduce the collective behavior of hundreds of cells in line with time-lapse microscopy [22 , 34] . Our major findings point towards previously uncharacterized features in aggregation , both observable in simulations and data . First , the transition to the collective is exactly pinpointed by a sharp rise in the spatial information of the cells during streaming . Second , to quantify the nature of the transition , we used fluctuations around the mean velocity , allowing us to distinguish between a hierarchically driven , top-down ( external gradient from leader cells ) and an emergent , self-organized , bottom-up ( all cells are equal ) process . Third , similar to second-order phase transitions in physical systems , the streaming phase shows signatures of criticality using finite-size scaling arguments . As a result , there is no intrinsic length scale , allowing cells to communicate with each other over large distances “for free” ( i . e . , only based on local cell—cell coupling ) . The control parameter is cell density , affecting the cell—cell coupling via cAMP secretion and sensing . Our work provides further insights into the process of cell aggregation . By means of our multiscale model , we were able to answer why cells emit cAMP in pulses . Albeit short lived , a pulse creates a steeper spatial cAMP gradient than continuous secretion ( assuming that the total amount of emitted cAMP is the same in both cases ) . Moreover , we noticed that so-called cAMP “waves” are likely not actual macroscopic traveling waves because of strong dissipation and diffusion . In contrast , cells are exposed to short-range cAMP pulses , which need to be relayed from one cell to the next before they dissipate . Although cAMP waves from microfluidic devices were used to study the cellular response to positive ( incoming wave ) and negative ( passing wave ) gradients , they may not represent natural stimuli [23 , 24] . Hence , cells may not have to solve the traditional “back-of-the-wave” problem but instead have to decide which pulse to follow . However , this difficulty is eased as cells secrete cAMP from their rear [39] . Indeed , experiments of constitutively expressed adenylyl cyclase show defective streaming [59] . Our multiscale model captures true emergence , which is generally not included in previous models of Dictyostelium aggregation . Models of wave propagation and spiral wave patterns go back as early as the 1970s [60] , but generally these models did not include cell motility ( but see [61] for an exception ) . More elaborate models from the 1990s focused on actual aggregation [16 , 20 , 29–31 , 46] . These were followed by the biologically more detailed LEGI [27 , 28] and Meinhardt [26 , 37] models to address the single-cell response to chemoattractant gradients . More recently , the FitzHugh—Nagumo model was adopted to explain the pulsing and synchronization of multiple cells ( see Supporting information for a comparison ) [34 , 35] , although early attempts to understand cAMP oscillations and the signal relay were already conducted in the 1980s [20] . Furthermore , hybrid models were proposed [62] . However , none of these models started from a detailed , spatiotemporal , single-cell model and was able to quantify the cell—cell correlations , type of order , and exact transition point for achieving collective behavior . When dealing with complex biological phenomena , there are necessarily limitations in the deduced models and acquired data . To assess criticality via finite-size scaling , ideally cell density is varied by orders of magnitude . However , this is often difficult to achieve in biological systems and depends on experimental conditions . On the one hand , if cell density is much lower than about 1/3 ML , cells do not aggregate [34] ( although lower density aggregation was achieved in a different experimental setup [63] ) . On the other hand , if the cell density is higher than 1 ML , experiments would need to be conducted in 3-D with major technical difficulties . Despite the approximations , our model allows the identification of the key ingredients for certain observed behavior . For instance , an earlier version of our model showed some level of aggregation but no finite-size scaling . By investigating this shortcoming , we noticed that streams were too narrow because of nearly negligible volume exclusion . However , quasi—one-dimensional streams restrict cell movement and suppress criticality , reminiscent of the missing disorder—order phase transition in the 1-D Ising model according to the Mermin—Wagner theorem [64] . ( Note that the 2-D Ising model is a borderline case , but it is still possible to formally define a phase transition according to Kosterlitz and Thouless [65] . ) In our simulations , only when volume exclusion is increased and streams become broader does critical-like behavior emerge ( see also Discussion in [47] ) . In an attempt to unify wide ranging biological phenomena , short-range interactions may play similar roles in cell collectives ( Dictyostelium , neurons , biofilms , embryos , tumors ) [10 , 66 , 67] and animal groups ( such as bird flocks ) [8 , 9 , 68–70] . Interestingly , many different cell types communicate by pulsing ( spiking ) , including neurons and bacteria [71] . Operating at criticality ( i . e . , the tipping point between order and disorder ) may allow cells to be maximally responsive; to communicate robustly over long distances; to act as a single , coherent unit; and to make decisions on , e . g . , when and where to aggregate . In the future , it would be fascinating to conduct aggregation experiments in 3-D environments and to study the collective response to perturbations such as obstacles , changes in temperature , and exposure to toxins .
The intracellular cAMP dynamics are described by the FitzHugh—Nagumo model , a classical model to reproduce neuronal spiking that was previously adopted to describe excitability in Dictyostelium [34 , 35] . Degradation of intracellular cAMP is achieved by phosphodiesterase regA , which is negatively regulated by extracellular concentration of cAMP ( by means of extracellular signal—regulated kinase ERK2 [38] ) . Secretion of cAMP from the cell rear [38 , 39] is strictly coupled to its intracellular concentration: if the extracellular cAMP concentration is below a threshold value , cells exhibit a constant small leakage of cAMP , but a temporary high concentration of cAMP is released during pulses of intracellular cAMP once above the threshold . If the extracellular cAMP concentration is kept above this threshold , the cell becomes a sustained oscillator . Extracellular cAMP is degraded by the phosphodiesterase PDE [72] . This model correctly captures the relay of the signal and the sustained pulsing observed in Dictyostelium ( see Supporting information for a detailed explanation ) . To reproduce the dynamics of thousands of cells , we simplified further the representation given by the detailed model . We assumed that cells are point-like objects that secrete cAMP maximally at their rear . Specifically , spatial propagation of cAMP was modeled as an exponential decay with a constant of 0 . 1 μm–1 ( within a factor of 2 of the value extracted from the detailed model simulations ) . The spatiotemporal concentration profiles are rescaled according to the cosine of the angle with the opposite-to-motion direction; secretion becomes zero at 90° ( lateral secretion ) and is set to zero for all the frontal part of the cell . ( The above-mentioned fine tuning of the exponential decay constant may be a result of this rescaling approximation or may reflect the fact that the cell—cell coupling is a key parameter for critical-like behavior . ) We set a concentration threshold c1 to determine if a given cell will emit a pulse or just leak cAMP , and a gradient threshold ∇c2 determines if the cell will move randomly or follow the local cAMP gradient . As for the detailed model , every cell undergoes a refractory period of 6 min after firing , during which it keeps the same motion it had during pulsing . To reproduce volume exclusion , cells cannot be closer to each other than 3 μm ( this rule is overwritten later in simulations , when cells are densely packed and likely superimpose ) . To drastically speed up simulations , the algorithm is written without explicit modeling of diffusion of cAMP in space; instead , it computes how much cAMP every cell senses and what their spatial gradients are by considering positions of cells with respect to each other . This implementation is able to reproduce aggregation of thousands of cells . More specifically , n = 1 , 000 cells were considered at experimental density of about one monolayer ( 1 ML = 6 , 600 cells/mm2 ) . For the other simulations of n = 600 , 800 , and 1 , 200 , the total area ( of 389 x 389 μm ) was fixed and density varied accordingly . See Supporting information for a detailed explanation . The pair-correlation function was computed as described in [73] , given by g ( r ) =AN ( N−1 ) 12πra∑i≠jNδ ( r−rij ) ( 3 ) where A is the total area of the image considered , N is the number of cells , r is the radius of a ring , and a is the discretization constant . In case of a random distribution g ( r ) takes a value of 1 on average ( similar to blue trace in Fig 1D , inset ii ) , while in case of particle clustering , g ( r ) becomes greater for small distances ( as for the red trace in the same panel ) . Spatial information of an image of cells was calculated in Fourier space of wave numbers based on the formalism described in [44] . All images were binarized ( by means of MATLAB thresholding algorithms graythresh and im2bw for the case of experimental images ) . After that , 2-D images were converted in 3-D binary matrices in which the third dimension has a 1 corresponding to the pixel intensity ( thus , in this case , since the starting images were binary , the 3-D matrix has a 1 at level 0 if that pixel is black and at level 1 if it is white ) . This guaranteed that all images had the same histogram , provided that they initially were of the same size . For the case of uncorrelated pixels , all Fourier coefficients Pi are considered independent and Gaussian distributed . Image entropies were then calculated as: HkS=−2N∑iPi log2 Pi ( 4 ) where the probability density function P is Gaussian distributed with zero mean and variance calculated from the sum of the pixel intensities . Hks is computed by dividing the function into bins of width σ/100 and summing Pi log2 Pi from −10σ to 10σ . Fourier transformation was then applied to the image . The real and imaginary part of the Fourier coefficients were then considered to compute IkS=∑i ( −log2 PiR−log2 PiI ) ( 5 ) where PiR and PiI refer to the real and imaginary part of coefficient i . The sum was calculated by considering bins of width σ/100 around the values assumed by the Fourier coefficients . k -space spatial information kSI was finally calculated as kSI = HkS − IkS . For a primer on information theory , see Supplementary information . To calculate the connected correlations , local averages of the velocities were subtracted from cell velocities . For every cell , we considered the average movement of all cells in its neighborhood up to a certain maximal distance rc and computed the correlations between the cell in the center and all the cells belonging to its neighborhood . We repeated this procedure for every cell in our image . In this way , we were able to decrease the “bulk" velocity component in the fluctuations while keeping a continuous partition of the image ( which we would have lost in case of rigid partition of the image in smaller squares ) and without preassigning the final position of the aggregation center . In order to understand better the influence of this partitioning on the calculation of the connected correlations , we repeated the same procedure for different radii . Specifically , if L is the image dimension , we set rc equal to L/2 , L/4 , L/6 , L/8 , and L/10 , with L/6 appearing to be the best choice in terms of the tradeoff between avoiding overestimation of correlations and number of cells in the neighborhood for good statistics in the simulated data . For the analysis of experimental data , L/2 , L/3 , L/4 , L/5 , L/6 , and L/8 were considered and L/4 was chosen , reflecting again the tradeoff between good statistics of a noisy dataset and a small overestimation of correlations . To plot the susceptibility , we estimated the nearest-neighbor distance , computed for every frame as the average of the nearest-neighbor distances for all cells . Time-lapse movies were obtained similar to the protocol in [22 , 34] . Axenic Dictyostelium cells expressing the Epac1camps FRET sensor were starved for 4–5 h and then plated on hydrophobic agar for imaging . Sixteen fields of view from a microscope were combined ( 1 . 2 x 1 . 2 mm2 ) , resulting in the recording of thousands of cells in a wide field ( inverted epifluorescence microscope [TE300 , Nikon] ) . To allow high-precision tracking of individual cells in a dense cell population , a different fluorescent marker , mREPmars ( TRED ) , was expressed and mixed with unmarked cells so a subpopulation of cells could be tracked ( 10% TRED cells ) . See Supporting information for further details . Images of TRED channels were segmented by using the MATLAB function imextendedmax , which outputs a binary image given by the computation of the local maxima of the input image . The centroids positions were then computed from this mask by means of the regionprops function . The tracking of individual cells was done by considering the centroid positions for different times . For every time t , the nearest-neighbor centroid at time t+ 1 was found , and the trajectory was accepted if the distance between the two positions was smaller than the average cell size . | Cells are often coupled to each other in cell collectives , such as aggregates during early development , tissues in the developed organism , and tumors in disease . How do cells communicate over macroscopic distances much larger than the typical cell—cell distance to decide how they should behave ? Here , we developed a multiscale model of social amoeba , spanning behavior from individuals to thousands of cells . We show that local cell—cell coupling via secreted chemicals may be tuned to a critical value , resulting in emergent long-range communication and heightened sensitivity . Hence , these aggregates are remarkably similar to bacterial biofilms and neuronal networks , all communicating in a pulselike fashion . Similar organizing principles may also aid our understanding of the remarkable robustness in cancer development . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"cell",
"physiology",
"cell",
"motility",
"medicine",
"and",
"health",
"sciences",
"dictyostelium",
"fruiting",
"body",
"dictyosteliomycota",
"microbiology",
"simulation",
"and",
"modeling",
"fluorophotometry",
"physiological",
"processes",
"developmental",
"biology",
"dictyostelium",
"spore",
"cells",
"experimental",
"organism",
"systems",
"microbial",
"growth",
"and",
"development",
"research",
"and",
"analysis",
"methods",
"cell",
"movement",
"spectrum",
"analysis",
"techniques",
"fluorescence",
"resonance",
"energy",
"transfer",
"microbial",
"physiology",
"slime",
"molds",
"cell",
"communication",
"protozoan",
"models",
"dictyostelium",
"growth",
"and",
"development",
"chemotaxis",
"spectrophotometry",
"cell",
"biology",
"physiology",
"secretion",
"biology",
"and",
"life",
"sciences",
"dictyostelium",
"protists",
"organisms"
] | 2017 | A critical-like collective state leads to long-range cell communication in Dictyostelium discoideum aggregation |
Mitochondrial processing peptidases are heterodimeric enzymes ( α/βMPP ) that play an essential role in mitochondrial biogenesis by recognizing and cleaving the targeting presequences of nuclear-encoded mitochondrial proteins . The two subunits are paralogues that probably evolved by duplication of a gene for a monomeric metallopeptidase from the endosymbiotic ancestor of mitochondria . Here , we characterize the MPP-like proteins from two important human parasites that contain highly reduced versions of mitochondria , the mitosomes of Giardia intestinalis and the hydrogenosomes of Trichomonas vaginalis . Our biochemical characterization of recombinant proteins showed that , contrary to a recent report , the Trichomonas processing peptidase functions efficiently as an α/β heterodimer . By contrast , and so far uniquely among eukaryotes , the Giardia processing peptidase functions as a monomer comprising a single βMPP-like catalytic subunit . The structure and surface charge distribution of the Giardia processing peptidase predicted from a 3-D protein model appear to have co-evolved with the properties of Giardia mitosomal targeting sequences , which , unlike classic mitochondrial targeting signals , are typically short and impoverished in positively charged residues . The majority of hydrogenosomal presequences resemble those of mitosomes , but longer , positively charged mitochondrial-type presequences were also identified , consistent with the retention of the Trichomonas αMPP-like subunit . Our computational and experimental/functional analyses reveal that the divergent processing peptidases of Giardia mitosomes and Trichomonas hydrogenosomes evolved from the same ancestral heterodimeric α/βMPP metallopeptidase as did the classic mitochondrial enzyme . The unique monomeric structure of the Giardia enzyme , and the co-evolving properties of the Giardia enzyme and substrate , provide a compelling example of the power of reductive evolution to shape parasite biology .
The acquisition of the mitochondrial endosymbiont and its evolution into the mitochondrion were key events in the evolution of eukaryotes [1] . During this process , most of the protomitochondrial genome was either lost or transferred to the nucleus of the host cell [2] . As a consequence , most mitochondrial proteins are host-nuclear encoded and must be specifically targeted to the organelle where they function . In the best understood system , N-terminal extensions attached to mitochondrial matrix proteins are specifically recognised by receptors on the mitochondrial surface , and the preproteins are subsequently imported by translocases of the outer and inner mitochondrial membranes [3] . A final step in the import process is the removal of the N-terminal extension , by the mitochondrial processing peptidase ( MPP ) [4] , to prevent it from interfering with protein function and/or stability [5] . The MPP comprises a catalytic βMPP subunit that binds a zinc cation using amino acid residues of the conserved motif HXXEHX76E [6] , and a regulatory αMPP subunit with a flexible glycine-rich loop that is important for substrate recognition [7] . The two subunits together form a negatively charged cavity that accommodates and immobilizes presequences during processing [6] . The activity of MPP thus requires the cooperative action of both subunits; neither subunit is functional alone [6] , [8] . Mitochondrial targeting presequences are characterized by the ability to form a positively charged amphipathic alpha helix , but otherwise show little primary sequence conservation [6] . Their most prominent common feature is the presence of a cleavage motif , which determines the peptide bond to be cleaved by the processing peptidase . The cleavage motif includes a positively charged residue , typically arginine , at the -2 or -3 position from the cleavage site ( P2 or P3 ) , which is followed by hydrophobic ( P1′ ) and hydrophilic ( P2′ , P3′ ) residues [9] . Mutational analyses indicate that the P2 ( P3 ) arginine plays a key role in the recognition of the processing site by MPP and interacts with the glutamate of the βMPP active site [9] . In addition , there are one or more basic amino acid residue ( s ) N-terminally distal from the processing site that bind to acidic residues of the MPP cavity and stabilize the substrate-MPP complex [10] . Mitosomes and hydrogenosomes are highly reduced versions of mitochondria that are found in diverse parasitic or free-living unicellular eukaryotes inhabiting oxygen-poor or intracellular niches [1] . The organelles found in human parasites Giardia intestinalis and Trichomonas vaginalis lack a genome so all of their proteins are encoded by the nuclear genome and must be imported [1] . Some hydrogenosomal and mitosomal proteins have N-terminal extensions that are reminiscent of the presequences that direct proteins into mitochondria and they contain distinguishable cleavage motifs [11] , [12] . This suggests that the Giardia and Trichomonas organelles may also contain an MPP-like enzyme . A single gene coding for a putative processing peptidase has been found in the genome of G . intestinalis [13] and the gene product has been shown to localize in mitosomes [14] . The primary structure of GPP is highly divergent from mitochondrial homologues , with only 13 . 1% identity and 29 . 7% similarity to the βMPP of Saccharomyces cerevisiae . A single gene for a βMPP homologue ( 20 . 9% identity and 42 . 9% similarity to S . cerevisiae βMPP ) was also recently identified in the genome of T . vaginalis [15] . In this case , functional data were presented suggesting that the hydrogenosomal processing peptidase ( βHPP ) functioned as a homodimeric enzyme [15] . No αMPP homologue was detected , although a protein rich in glycine amino acid residues ( GRLP ) , that shares a limited similarity with the glycine-rich loop of αMPP , was located to T . vaginalis hydrogenosomes . However , GRLP was reported not to stimulate βHPP activity in vitro [15] . The progenitor of MPP was probably a monomeric α-proteobacterial peptidase , similar to the recently described Rickettsia prowazekii processing peptidase ( RPP ) [16] . During the evolution of mitochondria , gene duplication and subunit specialization gave rise to the heterodimeric α/βMPP , which is now present in the mitochondrial matrix or integrated as the core I and II subunits of the cytochrome bc1 complex in the inner mitochondrial membrane [6] . The single subunit structure of GPP and HPP [15] could thus reflect retention of the ancestral form of organization , or reductive evolution from the classic MPP heterodimer . It has also been suggested that the Giardia protein may have had a separate origin by lateral gene transfer from a bacterium other than the mitochondrial endosymbiont [13] . Here we show that GPP functions as a monomer consisting of a single βMPP homologue while HPP , like classical MPP , is fully active only upon heterodimerization of an α and β subunit . Based upon phylogenetic and functional analyses we infer that the unique monomeric structure of the Giardia mitosomal processing peptidase GPP , is the result of reductive , substrate-driven evolution from a heterodimeric progenitor enzyme .
To investigate the origins of the MPP-like proteins of Giardia and Trichomonas and the Trichomonas GRLP we carried out a phylogenetic analysis . As these proteins are heterogeneous for their amino acid compositions , and because a failure to accommodate such heterogeneity can lead to incorrect trees [17] , we used a recently described node-discreet-compositional-heterogeneity method to analyze the data [17] . A heterogeneous model comprising 10 composition vectors was found sufficient to produce data of similar composition to the original sequences , as judged by Bayesian posterior predictive simulation [17] ( Fig . S1 ) . Phylogenetic analyses using this model support the hypothesis that GPP , βHPP and βMPP share a common origin . This result contrasts with a previous analysis , using a poorly fitting composition homogeneous model , when GPP was reported to have no phylogenetic affinity with either MPP or the α-proteobacteria [13] . The position of the GPP among βMPP , together with the presence of the catalytic motif HXXEHX76E , are consistent with the protein being a βMPP-like peptidase ( βGPP ) , and not an αMPP-like protein as currently annotated [13] . Importantly , these data , together with the absence of an αMPP-like protein coding sequence on the Giardia genome , support the hypothesis that the single subunit structure of GPP results from reductive evolution including loss of an αMPP-like subunit . The alternative possibility , that the simple GPP structure reflects retention of the ancestral form of organization , is not supported by our analyses . Our results suggest that αMPP and βMPP probably arose once by a primordial gene duplication at the base of eukaryotes , and that all MPP-like proteins share common ancestry with single subunit enzymes from α-proteobacteria , consistent with an origin from the mitochondrial endosymbiont ( Fig . 1A ) . Notably , our analyses show that the T . vaginalis GRLP is part of the αMPP clade , suggesting that , contrary to previous claims [15] , T . vaginalis may possess a functional homologue ( GRLP ) of αMPP ( henceforth αHPP ) . To investigate the functionality of the βGPP , βHPP and αHPP-like proteins , we expressed them in E . coli . The recombinant βGPP processed the N-terminal extensions of Giardia mitosomal ferredoxin ( Gifdx ) and the iron-sulphur cluster scaffold proteins ( GiiscU and GiiscA ) . The processing activity was demonstrated as a shift in the substrate gel mobility and the cleavage sites were identified by N-terminal amino acid sequencing of the cleaved products ( Fig . 2 ) . The activity of the recombinant βGPP was inhibited by the chelator EDTA , and activity was also lost when the first glutamate of the HXXEHX76E motif was mutated to glutamine ( Fig . 3 ) . These data indicate that the βGPP is an active metallopeptidase with a similar cleavage mechanism to MPP [6] . Like the rickettsial homologue of MPP [16] , βGPP was active as a monomer , which was demonstrated by size exclusion chromatography of recombinant βGPP as well as by analysis of βGPP from a mitosome-rich fraction separated on a sucrose gradient under native conditions ( Fig . 1B , C ) . Importantly , kinetic parameters of monomeric βGPP ( Vmax = 0 . 27 µM/min; Km = 8 . 4 µM , Fig . S3 ) were comparable to those published for the heterodimeric MPP of Neurospora crassa [8] . It has recently been suggested that the T . vaginalis HPP functions as a homodimer of two identical βHPP subunits [15] , so we investigated the activity of βHPP with- and without αHPP . Unlike for βGPP , no activity for βHPP alone could be detected by gel shift assay ( Fig . 1D ) , but a small amount of activity was observed when a highly sensitive fluorometric assay was used [15] ( Fig . 1E ) . However , the processing activity measured by this assay increased by almost two orders of magnitude when the βHPP was associated with the αHPP-like protein , indicating that–like classic MPP–the T . vaginalis HPP functions most efficiently as a heterodimer ( Fig . 1E ) . To further investigate the structure-function relationships of the GPP , HPP and MPP , we screened in silico the G . intestinalis and T . vaginalis proteomes for putative mitosomal and hydrogenosomal N-terminal presequences ( Tables S1 and S2 ) , which were then analyzed for structural elements known to mediate substrate-MPP interactions . In particular , we searched for the positively charged residues proximal to the cleavage site ( P2 or P3 ) , and those which are N-terminally distal from the processing site . The distance between the proximal and distal group was defined to be at least 3 amino acid residues [18] , [19] . Giardia mitosomal presequences were predicted in three of nine putative mitosomal proteins ( Table S2 ) . All of these presequences possess the proximal P2 arginine within a conserved cleavage motif [ ( ARV ) R ( F/L ) ( L/I ) T] , but the distal positively charged residues are absent ( Table S2 ) . The lengths of the Giardia mitosomal presequences that have been experimentally verified are 10 , 12 and 18 amino acid residues . The majority of the in silico predicted Trichomonas hydrogenosomal presequences ( 147 ) resemble the Giardia pattern; having a length of 4 to 21 amino acid residues , possessing a P2 arginine within a cleavage motif , and lacking the distal positively charged residues . However , we also detected 79 putative hydrogenosomal presequences , of 10 to 24 amino acids , that–like classic mitochondrial sequences–do contain distal arginines or lysines at position P6–P22 . To compare the specificities of the βGPP , α/βHPP and yeast α/βMPP in vitro , we tested their activity on a selection of mitosomal , hydrogenosomal and mitochondrial substrates ( Fig . 2 ) . The βGPP cleaved only its own mitosomal substrates . By contrast , the α/βHPP cleaved the hydrogenosomal presequences , and the presequences of mitosomal ferredoxin and two mitochondrial substrates . The yeast α/βMPP processed all of the mitochondrial substrates and the two mitochondrial-like hydrogenosomal substrates that possess distal positively charged residues . We also tested whether we could make chimeric peptidases using a combination of hydrogenosomal and mitochondrial subunits . Interestingly , while the yeast αMPP did not interact with the Trichomonas βHPP , Trichomonas αHPP was able to form a heterodimer with yeast βMPP . However , this heterodimer did not cleave mitochondrial or hydrogenosomal substrates under our experimental conditions ( data not shown ) . To gain further insights into the structure-function basis of their different substrate spectra , we modelled each of the different proteins ( Fig . 4 and Fig . S4 ) , using the yeast MPP structure as a guide [10] . For yeast MPP , the substrate is first recognized by the glycine-rich loop of αMPP [6] , [7] and then moved to the active site of βMPP which interacts with the substrate cleavage motif including the proximal arginine . The distal positive residues of the presequence help to stabilize the substrate-MPP complex by binding to negatively charged residues within the large polar cavity formed by the α/βMPP subunits [10] . The part of the substrate-binding cavity formed by βMPP thus displays an evenly distributed negative charge to accommodate both proximal and distal positively charged residues of mitochondrial presequences . The αMPP interacts only with the distal positive residues of longer ( >20 amino acid residues ) mitochondrial presequences [9] , [10] . As the βGPP functions as a monomer we predict that its substrates , including the proximal arginine , interact directly with the negatively charged region of its catalytic site ( Fig . 4 ) . The rest of the predicted βGPP cavity is , unlike βMPP , positively charged , although its predicted overall fold structure still resembles that of βMPP ( Fig . S4 ) . The difference in βGPP charge distribution is compatible with the absence of distal positively charged residues in the mitosomal presequences , and , along with the absence of an αMPP-like subunit , may explain the inability of βGPP to process mitochondrial-type presequences . The simplicity of GPP is consistent with the highly reduced function of mitosomes and likely reflects ( i ) the paucity of proteins that are targeted to this organelle when compared with mitochondria and ( ii ) lack of N-terminal cleavable presequences in the majority of mitosomal proteins , including βGPP itself . As shown above ( Table S2 ) , only nine mitosomal proteins have been identified so far and these are involved either in organelle biogenesis ( Gipam18 , GiHsp70 , GiCpn60 , GPP ) or the formation of Fe-S clusters ( GiiscS , GiiscU , GiiscA , Gigrx , Gifdx ) , which is currently the only known mitosomal function for G . intestinalis . Of these , seven are targeted to mitosomes in the absence of a detectable N-terminal targeting signal and thus function independently of GPP . Other than these , no other homologues of mitochondrial proteins have so far been identified in the genome of G . intestinalis [13] . The T . vaginalis HPP represents an intermediate stage between GPP and MPP in terms of charge distribution and enzymatic activity . Thus , it can process presequences with- or without distal positive residues , but can only cleave the shorter mitochondrial presequences ( Fig . 2 ) . The presence of mitochondrial type presequences on hydrogenosomal proteins is consistent with the retention of the αHPP , which is likely involved in their recognition via its glycine-rich loop and/or their docking at the cleavage site . Our phylogenetic and functional analyses show that the Giardia GPP is a striking example of reductive evolution from a heterodimeric to a monomeric enzyme , with properties resembling the putative ancestral α-proteobacterial enzyme , rather than the highly specialized MPP heterodimer found in well characterized mitochondria . While the principal selective pressure for the evolution of the processing peptidases is probably their ability to efficiently process substrates , the differences in the properties of the substrate presequences may also reflect the mode of their translocation across the organelle membranes [20] . In mitochondria from species across the phylogenetic tree [21] , the positive residues of N-terminal presequences are recognized by the outer membrane TOM system and then the inner membrane translocase complex TIM23 [3] . Interestingly , no receptors ( Tom20 , Tom 22 , Tom70 ) or components of the translocation channel of the TOM complex ( Tom40 , Tom5 , Tom6 , Tom7 ) have so far been identified for G . intestinalis [13] or T . vaginalis [22] . Putative core components of the TIM23 translocase ( Tim23 , Tim17 ) as well as Pam18 involved in protein transfer to the matrix have been found in T . vaginalis , but only Pam18 was found in G . intestinalis [14] . It thus appears that reductive evolution of the organelles has dramatically affected both the processing peptidases and the protein import pathway [21] , with important implications for general models of mitochondrial biosynthesis , structure and function .
Complete sequences of βGPP , βHPP , and βMPP were aligned with Muscle [23] to calculate sequence identity and similarity values . MPP , GPP and HPP sequences were aligned with Muscle [23] and analysed with Gblocks [24] to remove ambiguously aligned sites . Bayesian phylogenetic analyses were conducted using P4 ( http://www . nhm . ac . uk/research-curation/projects/P4/index . html ) . The optimal substitution model for Bayesian analyses was identified by ProtTest [25] ( WAG+Gamma ) , a polytomy prior [26] , and one or more base composition vectors , which were free to vary during the chain under the NDCH model [17] . MCMC chains were run for 1 , 000 , 000 generations , sampling trees and parameters every 200 generations . Model parameter proposal tuning values were determined using the P4 “autoTune” method . The burn-in was identified using the method of Beiko and co-workers [27] . The base composition component of the model was tested by simulation of the base composition χ2 statistic [17] at each sampling point , resulting in a posterior predictive distribution [28] against which the statistic of the original data could be tested using tail-area probability . Composition vectors were successively added until adequate fitting of the observed data to the model was identified ( see Fig . S1 ) . The βGPP ( NCBI accession: XP_001707100 ) , αHPP ( XP_001276882 ) and βHPP ( XP_001316822 ) subunits and their substrates were expressed with hexahistidine tags in E . coli . An α/βHPP heterodimer was assembled from βHPP-His and non-tagged αHPP subunits by incubation of lysates of E . coli expressing the respective proteins for 30 min on ice in 20 mM Tris , 20 mM NaCl ( pH 8 . 6 ) , 1 mM MnCl2 . All recombinant proteins were purified by nickel column chromatography ( HiTrap Chelating ) under native ( βGPP-His , αHPP-His , βHPP-His , and α/βHPP-His ) or denaturing ( substrate proteins ) conditions . An α/βMPP heterodimer was prepared as published [18] . The GPP reactions were carried out in 20 mM Tris ( pH 8 . 0 ) , 100 mM NaCl , 1mM MnCl2 , 30 min at 37°C , the HPP reactions in 20 mM Tris-HCl ( pH 8 . 6 ) , 20 mM NaCl , 2 mM MnCl2 , 30 min at 37°C and activity of MPP was determined in 50 mM HEPES ( pH 7 . 4 ) , 20 mM NaCl , 1 mM MnCl2 , 30 min at 30°C . To identify the cleavage sites , all substrates processed by the three proteases were subjected to N-terminal protein sequencing by Edman degradation . The kinetics of GPP was determined using the method published by Arretz and co-workers [8] . For determination of the activity of the HPP subunits , purified αHPP-His and βHPP-His were incubated on ice for 30 min either alone , or mixed together with 1 mM MnCl2 . After addition of TviscU , the reaction was allowed to proceed at 37°C for 60 min . The specific activity of HPP with a fluorescent substrate based on the presequence of TvAK [Abz-MLST LAKRF AY ( NO2 ) GKKDRM] ( Bachem , Switzerland ) was measured at 420 nm , with an excitation wavelength of 315 nm ( Infinite M200 , Tecan ) . A pre-calibrated Superdex 200 column was used to determine the molecular mass of E . coli produced GPP , under native conditions . Affinity purified βGPP-His in buffer of 50 mM CHES ( pH 9 . 5 ) , 150 mM NaCl was loaded on the column and washed ( 0 . 5 ml/min ) , collecting 1 ml fractions . Protein-containing fractions were assayed for βGPP activity . The molecular mass of GPP expressed in G . intestinalis with a hemagglutinin ( HA ) tag was estimated under native conditions by sucrose gradient centrifugation [15] . The mitosome-enriched fraction was isolated from a G . intestinalis homogenate using a published method [14] . The proteins in the mitosomal-enriched fraction were then separated on a calibrated sucrose gradient [15] . Fractions were analysed by immunoblot using anti-HA antibodies . Bands visualized by alkaline phosphatase were quantified by densitometry ( GS-800 Calibrated Densitometer , BioRad ) . An application based on the NetBeans Platform ( http://platform . netbeans . org ) was developed to search for proteins containing N-terminal hydrogenosomal and mitosomal presequences in the predicted T . vaginalis ( http://www . trichdb . org/trichdb/ ) and G . intestinalis ( http://www . giardiadb . org/giardiadb/ ) proteomes , respectively . Hydrogenosmal presequences were predicted based on two main parameters extracted from 21 known hydrogenosomal presequences: ( i ) the cleavage site motif , specified as RXF/ ( ILFSAGQ ) or R ( FNESG ) / ( ILFSAGQ ) ( the slash indicates the cleavage site and brackets mean one residue position ) , and the presequence start motif defined as ML ( STACGR ) or MTL or MSL . In addition , tryptophan was forbidden from the presequence , the maximum presequence length was optimized to 25 residues . Any presequences with overall negative charges were excluded ( the approximate presequence charge at pH7 was counted according to the Henderson-Hasselbalch equation using the following pKa values: N-terminus 8 . 0 , lysine 10 . 0 , arginine 12 . 0 , histidine 6 . 5 , glutamic acid 4 . 4 , aspartic acid 4 . 4 , tyrosine 10 . 0 , and cysteine 8 . 5 ) . The G . intestinalis proteome was searched for N-terminal presequences based on three experimentally verified mitosomal presequences of known mitosomal proteins [13] ( Table S2 ) . The parameters defined for the search were as follows: the cleavage site motif was defined as R ( FS ) / ( IL ) T , the presequence start motif as M ( SLT ) , the maximum presequence length was set up to 20 residues , tryptophan was forbidden from the presequence . A search using parameters for prediction of hydrogenosomal presequences did not reveal additional mitosomal protein candidates . The MODELLER program [29] version 9 . 2 was used to build 3-D models of αHPP , βHPP and βGPP . Alignments of the βGPP and βHPP with the βMPP ( pdbid 1HR6 ) [10] and of the αHPP with the αMPP ( pdbid 1HR6 ) [10] were carried out using the PROBCONS web service [30] and manually edited . The quality of the final model was checked using the ProCheck [31] and WhatCheck [32] programs . The electrostatic properties of the model were evaluated using APBS version 0 . 5 . 1 [33] . | In classic model organisms , cleavage of signals that are required to deliver nuclear-encoded proteins to mitochondria is mediated by an enzyme comprising two different subunits , called α or β , neither of which is functional by itself . Here , we have characterized a novel enzyme that functions in the mitosome , a highly reduced mitochondrion , of the pathogenic protist Giardia intestinalis . The Giardia enzyme is unique among eukaryotes because it has undergone reductive evolution to function efficiently as a single β-subunit monomer . We also show that the recent claim that the equivalent enzyme in the hydrogenosome , another type of reduced mitochondrion of the human parasite Trichomonas vaginalis , functions as a homodimer of two β-subunits , is not supported . The Trichomonas enzyme requires both an α- and a β-subunit to function most efficiently . Computational analysis of the Giardia and Trichomonas enzymes reveals that their structures and surface charge distributions have co-evolved to match the peculiar properties of the targeting signals that they process . The Giardia mitosome is an ideal model for studying the limits of mitochondrial reductive evolution and , because it makes cofactors that are essential for Giardia survival , is a potential therapeutic target for this important human parasite . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] | [
"evolutionary",
"biology",
"biochemistry/molecular",
"evolution",
"cell",
"biology",
"microbiology/parasitology"
] | 2008 | Reductive Evolution of the Mitochondrial Processing Peptidases of the Unicellular Parasites Trichomonas vaginalis and Giardia intestinalis |
A diverse suite of effector immune responses provide protection against various pathogens . However , the array of effector responses must be immunologically regulated to limit pathogen- and immune-associated damage . CD4+Foxp3+ regulatory T cells ( Treg ) calibrate immune responses; however , how Treg cells adapt to control different effector responses is unclear . To investigate the molecular mechanism of Treg diversity we used whole genome expression profiling and next generation small RNA sequencing of Treg cells isolated from type-1 or type-2 inflamed tissue following Leishmania major or Schistosoma mansoni infection , respectively . In-silico analyses identified two miRNA “regulatory hubs” miR-10a and miR-182 as critical miRNAs in Th1- or Th2-associated Treg cells , respectively . Functionally and mechanistically , in-vitro and in-vivo systems identified that an IL-12/IFNγ axis regulated miR-10a and its putative transcription factor , Creb . Importantly , reduced miR-10a in Th1-associated Treg cells was critical for Treg function and controlled a suite of genes preventing IFNγ production . In contrast , IL-4 regulated miR-182 and cMaf in Th2-associed Treg cells , which mitigated IL-2 secretion , in part through repression of IL2-promoting genes . Together , this study indicates that CD4+Foxp3+ cells can be shaped by local environmental factors , which orchestrate distinct miRNA pathways preserving Treg stability and suppressor function .
Regulatory T ( Treg ) cells [1] employ an arsenal of non-overlapping mechanisms to maintain immunological homeostasis at environmental interfaces [2] and internal organs [3] , preventing the development of hyper-inflammatory conditions [4] , [5] . The suppressive functions of Treg cells are crucial , without which fatal lympho- and myelo-proliferative autoimmune syndromes develop [6] . Restoring immunological homeostasis with regulatory T cell-based therapy may remedy some hyper-inflammatory conditions [7] . Regulatory T cells also restrict de novo responses to foreign antigens , limiting immunopathologies but sometimes at the cost of preventing natural , or vaccine-mediated , immunity [8] . In this context , temporarily disarming Treg functions may increase the efficacy of vaccines and immunity to infection . Elemental to any Treg-based therapeutic strategy is manipulating the appropriate Treg cells . Expression of the transcription factor forkhead box P3 ( Foxp3 ) in αβ+CD4+ lymphocytes activates and represses a suite of target genes [9] essential for Treg development and function . For this reason , Foxp3 expression is commonly used as a marker of Treg cells and is often used to compare Treg cells from a variety of different diseases . It has recently emerged that Foxp3+ Treg cells are heterogeneous and may be as diverse as the types of immune responses they regulate [10]–[14] . Foxp3+ Treg cells therefore represent a population of loosely related lymphocytes , still requiring greater molecular characterization . Foxp3+ cell development and function is intricately controlled transcriptionally by epigenetic modifications influencing gene accessibility [15] and post-transcriptionally by microRNAs ( miRNAs ) [16] . miRNAs have emerged as key regulators of innate and adaptive immune responses [17] and confer robustness and adaptability to cells in response to environmental fluctuation [18] . Disrupting canonical miRNA biogenesis by ablating the miRNA processing enzymes Dicer or Drosha in T cells [19]–[21] dysregulated T cell proliferation , differentiation , survival and cytokine production leading to a reduction in Foxp3+ cells and subsequent lethal inflammation [19] . Deletion of the entire miRNA repertoire specifically within Foxp3+ cells phenocopied Foxp3−/− mice with a loss of Treg function and the development of fatal autoimmunity [22] , [23] . These studies highlight the crucial role of miRNA-mediated gene regulation in Treg biology . However , which miRNAs are required for different Tregs and Treg-associated functions is poorly understood . Several miRNAs ( miR-21 , miR-31 , miR-24 and miR-210 ) [24] , [25] directly target Foxp3 in human T cells , regulating Foxp3 expression and Treg development . Additionally , Foxp3 activates miRNA-mediated mechanisms [25] to repress effector pathways , including suppression of SOCS1 via induction of miR-155 [26] . These studies indicate an intricate functional relationship between Foxp3 and miRNAs . Furthermore , Lu and colleagues [27] recently identified a role for miR-146a in regulating the expression of Stat1 , which is required for Treg-mediated control of Th1 responses . While such individual miRNA:target interactions are of interest , a single miRNA can target hundreds of mRNAs [28] , simultaneously regulating multiple pathways . We hypothesized that widespread miRNA-mediated regulation contributes to Foxp3+ cell diversity . To test this , we isolated Foxp3+ cells from mice chronically infected with Schistosoma mansoni , a parasitic helminth that invokes a polarised Th2 response , or Leishmania major , a parasitic protozoa controlled by Th1-mediated immunity . Microarray analysis revealed dramatically different gene expression profiles , confirming the heterogeneity of Foxp3+ cells . To investigate which miRNAs contribute to the observed gene expression differences , we first deep sequenced the small RNAome from these two Foxp3+ populations and identified several miRNAs that were significantly differentially expressed , relative to Treg cells taken from naïve mice . These miRNAs were analyzed further using our recently published in silico method [29] for predicting candidate ‘regulatory hubs’ . miR-10a was identified as the strongest such regulatory hub in L . major Foxp3+ cells , whereas miR-182 was the most critical in S . mansoni Foxp3+ cells . Gain and loss of function experiments in vitro and in vivo using primary Foxp3+ cells and Foxp3+ cells isolated from Th1 or Th2 inflamed tissue confirmed many of the predicted targets and functions for miR-10a and miR-182 . We further demonstrated that IL-4 up-regulates miR-182 , potentially through the transcription factor cMaf , which is also up-regulated by IL-4 . miR-182 critically restricted IL-2 production , possibly by its control of Bach2 [30] and Cd2ap [31] . We also showed that IL-12/IFNγ represses both miR-10a and its candidate upstream transcription factor Creb . Reduced miR-10a correlated with an increase in miR-10a target genes , Nr4a3 and Fbxo30 , which have previously been shown to control IFNγ . Collectively , this study supports the concept of heterogeneity , or plasticity , within the Foxp3+ pool and identifies candidate ‘regulatory hub’ miRNAs , miR-10a and miR-182 , which control IFNγ and IL-2 through essential gene programs .
Following infection with Schistosoma mansoni or Leishmania major , robust Th2 or Th1- responses develop [32] , [33] , accompanied by the recruitment of Foxp3+ Treg cells [34]–[39] . Genome-wide analysis of isolated Foxp3+ cells recruited to the liver of S . mansoni ( S . mansoni Foxp3+ ) or ear of L . major ( L . major Foxp3+ ) infected mice ( Fig . 1A ) identified distinct gene expression profiles , relative to Foxp3+ cells isolated from the spleen of uninfected mice ( Fig . 1B ) . Of the differentially expressed genes , 185 ( 11 . 6% ) were common between L . major and S . mansoni , whereas 441 ( 27 . 7%; S . m . ) and 967 ( 60 . 7%; L . m . ) were specific to each population ( Figure 1C ) , indicating that , with respect to gene expression , these Foxp3+ populations were substantially different from one another . The vast majority of the 185 common genes ( Table S1 ) were similarly regulated in each Treg population ( Figure S1 ) . Relative to Foxp3+ cells from the spleen of uninfected mice , L . major Foxp3+ cells upregulated several heat shock proteins ( Hsph1 , Hspa8 and Hspa1a ) , cytokine and chemokine-associated genes ( Il23r , Il33 , IL18R1 , Tgif1 , Cxcl10 , Rgs2 , Lph2 , Tnfaip3 ) and a range of transcriptional regulators ( Bcl6 , Mxi1 , Atf3 , Ror-α , Rel , Irf4 , Stat5a , Tfap2a ) ( Figure 1 D ) . S . mansoni-derived Foxp3+ cells , in contrast , upregulated genes associated with inhibition and killing ( Gp49a , Klrg1 , Gzma , Nkg7 , Lag3 , Tigit Cd200r1 and Cd200r1l , Trail ( Tnfsf10 ) ) , integrins and adhesion molecules ( Alcam , Epcam , Itga1 , Itgb8 , Selplg ( P-Selectin ) ) , cytokines and chemokines ( Csf1 , Il10r1 , Il12rb2 , Il18rap , Il1rn , Tgfβr1 , Socs2 , Ccl5 , Ccl1 , Ccl3l , Cxcl3 , Cxcr6 , Cxcr3 , Ccr1 ) , and different transcriptional regulators ( Tbx21 , Pparγ and Irf8 ) ( Figure 1E ) , many of which were also observed in a previous report [36] . To identify miRNAs that might contribute to the different expression profiles , we deep sequenced small RNA species from each of the three Foxp3+ populations ( S . m . , L . m . and Naïve ) and obtained 12–22 million reads in each sample ( Table . S2 ) . Within S . mansoni Foxp3+ cells , 31 miRNAs were differentially expressed ( p<0 . 05 ) ( Figure 2A and Table . S3 ) . HIF1α-inducible miR-210 and 2 poly-cistronic miRNAs , miR-183 and IL-2-inducible miR-182 , were the most significantly up-regulated ( Figure 2A ) . Seventeen miRNAs were differentially regulated ( p<0 . 05 ) in L . major Foxp3+ cells . Only one of these miRNAs , miR-100 , was up-regulated; while miR-32 and miR-10a were the two most significantly down-regulated ( Figure 2C and Table . S3 ) . Notably , down-regulation of miR-10a in L . major Foxp3+ cells was relative to ‘naïve’ Foxp3+ Treg cells , and not relative to naïve T cells , as recently reported [40] . Several miRNAs were differentially expressed in both Foxp3+ populations , including miR-151 , miR-30e , miR-15b , miR-132 , miR-342 , miR-10a and miR-32; however , not always in the same direction . For example , miR-132 , which regulates interferon-stimulated genes [41] , was ∼2-fold up-regulated in S . mansoni Foxp3+ cells , but ∼6-fold down regulated in L . major Foxp3+ cells . We next employed in silico Monte Carlo simulation analyses to identify which , if any , of the up- or down-regulated miRNAs in each Foxp3+ population are predicted to target significantly more of the down- or up-regulated mRNA transcripts , respectively , than expected by chance ( i . e . ‘regulatory hub’ miRNAs ) [29] . This approach identified miR-182 ( up-regulated in S . mansoni Foxp3+ cells ) as the strongest candidate regulatory hub of the network of down-regulated genes in S . mansoni Foxp3+ cells ( Figure 2B and Table S4 ) , and miR-10a ( down regulated in L . major Foxp3+ cells ) as the strongest candidate regulatory hub of the network of up-regulated genes in L . major Foxp3+ cells ( Figure 2D ) . To validate the predicted target genes of miR-182 and miR-10a , we isolated primary Foxp3+ cells ( predominantly nTreg cells ) , over-expressed or inhibited miR-182 or miR-10a using miRNA mimics or hairpin inhibitors , and measured miRNA and target mRNA expression . Transfection at >80% efficiency ( Figure S2 ) increased ( 20-fold ) or decreased ( 10-fold ) miR-182 using specific mimics or inhibitors ( Figure 3A ) . In contrast to naïve T cells [42] , expression of a previously reported miR-182 target , Foxo1 , was only marginally regulated by miR-182 in Treg cells failing to reach statistical significance ( Figure 3A , boxed ) . Of the 14 predicted targets in S . mansoni Foxp3+ cells ( Table S5 ) , 6 were significantly regulated ( >1 . 5 fold ) in response to miR-182 mimics or inhibitors . Similarly , miR-10a significantly regulated Hoxa1 , a previously validated miR-10a target [43] , along with 7 of the 11 genes in L . major Foxp3+ cells predicted to be targets of miR-10a ( Figure 3B and Table S5 ) . Collectively , using gain and loss of function for miR-182 and miR-10a in primary Foxp3+ cells , these data identify that miR-182 regulates 6 of the predicted genes identified in Th2-Treg cells and miR-10a regulates 7 of the predicted genes identified within Th1-Foxp3+ cells . To validate the functional significance of these miRNA:target interactions , and to determine whether differential expression of miR-182 and miR-10a was restricted to Foxp3+ cells from S . mansoni and L . major infections , we developed a Th1 and Th2-driven airway inflammation model . This system allowed us to eliminate pathogen influences , tissue-specific responses and any other factors that may have contributed to the observed Treg profiles observed above . Briefly , naïve T cells ( CD4+CD44loCD62LhiCD25− ) from congenic and transgenic C57BL/6 mice ( CD45 . 1+OTII+RAG2−/− ) were polarized in vitro under Th1 or Th2 conditions , secreting high levels of IFNγ or IL-5 respectively ( Figure 4A ) , and adoptively transferred into C57BL/6 CD45 . 2 Foxp3gfp mice . One-day prior to transfer ( d-1 ) and one and three days following transfer ( d1 and d3 ) , recipient mice received an intra-tracheal delivery of OVA into the lower airways ( Figure 4A ) . Adoptively transferred cells migrated to the lung and broncho-alveolar ( BAL ) spaces ( Figure 4B ) and caused peri-bronchial and peri-vascular inflammation ( Figure 4C ) . Antigen recall assays demonstrated that recipients of Th1 cells produced IFNγ and IL-10 ( Figure 4D ) and increased the expression of Inos , Mig ( Cxcl9 ) and Ip-10 ( Cxcl10 ) within the lung ( Fig . 4E ) . Mice that received Th2 cells secreted IL-4 , IL-5 and IL-9 ( Figure 4D ) and up-regulated Arg1 , Eotaxin ( Ccl11 ) and Gob5 ( Clca3 ) within the lung ( Figure 4E ) , characteristic of Th1 or Th2-mediated airway inflammation . CD4+Foxp3+ cells isolated from Th1- or Th2-inflammed lungs ( Figure 4F ) up-regulated Tbx21 , Gata3 , Foxp3 , Ctla4 , Gitr ( Tnfrsf18 ) , Il10rα , Ebi3 and Il10 with a small increase in Tgfβ in Th1-Treg cells only ( Figure 4G ) . As predicted , Foxp3+ cells from Th1 inflamed lungs down-regulated miR-10a with no change in miR-182 ( Figure 4 H ) , similar to Foxp3+ cells from L . major infected mice ( Figure 2 ) . Foxp3+ cells from Th2-inflamed lungs up-regulated miR-182 , with a marginal increase in miR-10a , similar to Foxp3+ cells from mice infected with S . mansoni ( Figure 2 ) . With the exception of Fosl and Cebpa , we also observed a very similar target gene expression profile in Th1-Treg or Th2-Treg cells isolated from the inflamed lung as compared to Treg cells from L . major or S . mansoni infected mice ( Figure 4I ) . These data support the notion that down regulation of miR-10a and up-regulation of miR-182 within Foxp3+ cells is associated with Th1 or Th2 biased immune environments , respectively . To test whether Th1 and Th2-associated Treg cells were functionally distinct from each other , we fluorescently-labeled OVA-specific Th1 or Th2 Teff ( CD4+CD44+Foxp3− ) cells isolated from Th1- or Th2-inflamed lungs , or naïve Teff cells as a control population , and co-cultured these cells with Th1- or Th2-Treg cells ( CD4+Foxp3+ ) from respective Th1 or Th2-inflamed lungs , or with Treg cells isolated from the opposing inflammatory environment in a series of ‘cross-over’ assays . In these assays , Th1-Treg cells potently suppressed Th1-Teff cells ( Figure S3A ) and Th2-Teff cells ( Figure S3B ) , whereas Th2-Treg cells only suppressed Th2 cells and not Th1 cells ( Figure S3C and S3D ) . We next tested whether down-regulated miR-10a and up-regulated miR-182 was functionally required for Th1- and Th2-Treg-mediated suppression , respectively . Th1-Treg cells isolated from the lungs of mice were transfected with miR-10a mimics ( Figure S3E ) , to overturn the down-regulated miR-10a observed in Th1-Treg cells ( Figure 4H ) . Following the observation that miR-182 was upregulated in Th2-associated Foxp3+ cells ( Figure 4H ) , Th2-Treg cells were transfected with miR-182 hairpin inhibitors ( Figure S3E ) . Mock-transfected Th1-Foxp3+ cells efficiently suppressed Th1 ( Figure 5A ) , Th2 ( Figure 5B ) and naive ( Figure 5C ) T cell proliferation . However , Th1-Treg cells transfected with miR-10a mimics were compromised in their ability to suppress Th1 cells ( Figure 5A ) and naïve T cells ( Figure 5C ) , but retained the ability to partially suppress Th2 cells ( Figure 5B ) . As a further control , we transfected Th1-Treg cells with miR-182 inhibitors , as miR-182 was not differentially regulated in Th1-Treg cells ( Figure 4H ) and this did not influence Th1-Treg mediated suppression of Th1 , naïve or Th2 cells ( Figure 5A , B and C ) . Th2-Treg cells were unable to suppress Th1 cells ( Figure 5D ) but were fully capable of suppressing Th2 ( Figure 5E ) and naïve T cells ( Figure 5F ) . Transfection with miR-10a mimics had no impact on Th2-Treg mediated suppression . However , Th2-Treg cells transfected with miR-182 inhibitors compromised their ability to suppress Th2 and naïve T cell proliferation , indicating that elevated miR-182 was required for Th2-Treg function . Treg cells isolated from the spleen of naïve animals were unable to control OVA-specific Th1 or Th2 cells ( Figure S4A and S4B ) , but were fully capable of suppressing naïve T cells ( Figure S4C ) . Transfection of Treg cells from naïve mice with miR-182 inhibitors or miR-10a mimics also compromised their suppressive capacity . Taken together , these data indicate that down-regulation of miR-10a is critically required for Th1-Treg cells to control Th1 cells and naïve T cells , while up-regulated miR-182 is required for Th2-Treg-mediated suppression of Th2 cells and naïve T cells , highlighting the divergence of these two Treg populations , while Treg cells from naïve mice were dependent upon both tightly regulated miR-10a and miR-182 . To determine the upstream factors that may contribute to miR-182 and miR-10a expression in Treg cells , we screened for transcription factor binding sites in the promoters of the primary transcripts of both miR-182 and miR-10a using Pwm-Scan ( as described in the methods ) . We identified putative binding sites in the miR-182 promoter for IL-4-regulated transcription factors ( TFs ) , including cMaf , and IL-12/IFNγ-regulated TFs , including Creb , in the miR-10a promoter ( Figure S5A ) . Concordant with the in-silico predictions , exposure of natural ( nTreg ) or in vitro generated inducible Treg ( iTreg ) cells ( Figure S5B ) to IL-4 , mimicking a Th2 environment , up-regulated cMaf ( Figure S5C ) , and miR-182 ( Figure S5E ) , similar to ex vivo Th2-Treg cells ( Figure 2 , and Table S1 ) . IL-12/IFNy treatment of nTreg and iTreg , mimicking the Th1 environment , down-regulated Creb ( Figure S5D ) and miR-10a ( Figure S5F ) in Treg cells , relative to naïve T cells , pheno-copying miR-10a expression in ex vivo Th1-Treg cells ( Figure 2 ) . Following recent studies indicating that Foxp3-mediated epigenetic modifications may be altered in Foxp3gfp-reporter mice [44] , [45] , we compared miR-182 and miR-10a expression in freshly isolated nTreg cells and in vitro generated iTreg cells from Foxp3rfp and Foxp3gfp-reporter mice , but did not observe any appreciable difference in miR-182 or miR-10a expression , relative to naïve T cells ( Figure S6 ) . CD2 , via Cd2ap and BACH2 , regulates IL-2 production through direct binding to the IL-2 promoter [30] , [31] . Following the observation that miR-182 targeted Cd2ap and Bach2 , and that IL-4 regulated miR-182 ( Figure S5 ) , we tested whether IL-4 influenced the expression of miR-182 , Bach2 , Cd2ap and subsequent IL-2 production . IL-4 treated nTreg or iTreg cells had reduced Bach2 and Cd2ap relative to naïve T cells or untreated Treg cells ( Figure S7A , S7B ) . We therefore assayed for IL-2 following IL-4 treatment , to determine whether IL-4-regulated miR-182 , and subsequent changes in Bach2 and Cd2ap had any influence on IL-2 responses . Il2 mRNA and protein levels were not altered following IL-4 treatment alone ( Figure S7E ) , however inhibition of miR-182 , with or without IL-4 treatment , led to a 50-fold induction of Il2 transcription and IL-2 secretion ( Figure S7E ) . These data indicate that miR-182 controls IL-2 production in Treg cells , possibly via Cd2ap and Bach2 , and that IL-4 re-enforces miR-182-mediated control of IL-2 . Previous reports have identified that Nr4a3 induces Foxp3 expression and represses IFNγ [46] . Following the observation that miR-10a targeted Nr4a3 we assayed for IFNγ following miR-10a over expression , with or without IL-12/IFNγ treatment . IL-12/IFNγ treatment alone induced IFNγ in Treg cells ( 4-fold , Figure S7F ) , similar to previous reports [47] , [48] , however IFNγ was increased 40-fold when combined with miR-10a over-expression ( Figure S7F ) . Interestingly , miR-10a over-expression alone also led to an increase in IFNγ ( 9-fold ) . Thus , type-2 regulated miR-182 and type-1-regulated miR-10a , respectively , contribute to the regulation of IL-2 and IFNγ responses in Th2- and Th1-Treg cells . To determine whether miR-10a and miR-182 was required for Treg survival , migration and control of Th1 and Th2-mediated inflammation in vivo , we designed a double adoptive transfer system ( Figure S8 ) . Briefly , Th1- or Th2-associated Foxp3+ Treg cells were isolated from Th1 or Th2–inflamed tissue , as above ( Figure 4 ) . A second recipient mouse received Teff ( OTII-Th1 or OTII-Th2 ) cells alone or a combination of mock-transfected Treg cells , miR-10a mimic transfected Th1-Treg cells with Th1-Teff cells , or miR-182-inhibitor transfected Th2-Treg cells with Th2-Teff cells . Following intra-tracheal delivery of OVA , similar percentages of transferred Treg cells were observed in the lung of recipient mice ( Figure 6A ) , indicating that all Treg cells experienced similar survival irrespective of transfection treatments . Significant numbers of inflammatory cells were recovered from the airspaces of mice receiving Th1 or Th2 cells ( Figure 6B ) , however the co-transfer of mock-transfected Treg cells significantly reduced the number of inflammatory cells . Co-transfer of Th1 cells and miR-10a mimic transfected Th1-Treg cells , or Th2 cells with miR-182-inhibitor transfected Th2-Treg cells failed to suppress inflammatory cell recruitment . The requirement for down-regulated miR-10a in Th1-Treg cells and up-regulated miR-182 in Th2-Treg cells was also reflected by uncontrolled IFNγ or IL-5 secretion in re-stimulated lymph nodes , compared to mice receiving mock-transfected Treg cells ( Figure 6C ) . Mock-transfected Treg cells potently reduced pulmonary pathology ( interstitial inflammation , mucus plugs and epithelial elongation ) , which was compromised when miR-10a or miR-182 was specifically deregulated in Th1- or Th2-Tregs , respectively ( Figure 6D ) . Taken together these studies highlight two diverse Foxp3 populations that develop to control Th1 or Th2 inflammatory events . The molecular programs in these Foxp3+ Tregs are in-part regulated by distinct upstream regulatory miRNA hubs , miR-182 and miR-10a , which target non-overlapping and essential genes within these diverse Foxp3+ populations .
In this study we identified distinct populations of Foxp3+ Treg cells recruited to Th1 or Th2 inflammatory environments expressing unique gene and miRNA profiles . Several genes and miRNAs were comparably regulated between the two subsets including miR-30e , miR-15b , miR-32 , miR-151 and miR-342 , with other miRNAs highlighting a clear divergence . For example miR-132 was significantly down regulated in Foxp3+ cells from Th1 rich surroundings ( −2 . 56 fold ) and up regulated in Foxp3+ cells in Th2 environments ( +2 . 09 fold ) . Using miRNA target prediction algorithms and Monte Carlo simulations we identified two miRNA regulatory hubs that target multiple genes contributing to the divergent gene expression profiles . Specifically , Th1 inflammation , following chronic L . major infection or acute Th1-induced inflammation , recruited Foxp3+ Treg cells that up-regulated a suite of genes regulated by miR-10a . In contrast , Foxp3+ cells isolated from Th2 environments following chronic S . mansoni infection or acute Th2-driven inflammation down-regulated a suite of genes under the control of miR-182 . These data support the notion that Foxp3+ cells are heterogeneous , or adaptable to their inflammatory environment [10]–[12] , [14] , [49] and provide an upstream molecular mechanism contributing to Foxp3+ heterogeneity . Previously , T-bet has been singled out as a co-transcription factor required for Treg cells to control anti-mycobacterial Th1 responses [12] . In our studies , Foxp3+ cells isolated from L . major infected tissue did not up-regulate T-bet , which may be explained by different infections , different stages of infection or different tissues studied . In support of the latter , and in agreement with the previous study , Foxp3+ cells isolated from Th1-inflamed lung tissue up-regulated T-bet ( ∼20-fold , Figure 4G ) , similar to pulmonary M . Tb . Foxp3+ cells . Interestingly , T-bet was greater than 200-fold up-regulated in Foxp3+ cells isolated from Th2-driven inflammation or from the liver of S . mansoni infected mice ( 2 . 6-fold ) . If Foxp3+T-bet+ cells are potent suppressors of Th1 responses , it is tempting to speculate that Foxp3+T-bet+ cells contribute to a dominant Th2 environment by potently suppressing Th1 responses . Similarly , Irf4 , a transcription factor involved in several T helper cell subsets [50] , [51] , was recently identified in Foxp3+ cells restraining Th2 responses . Irf4 however was not up-regulated in Th2-associated Treg cells isolated from schistosome infected mice , relative to Foxp3+ cells from the spleen of naïve mice , and was only slightly up regulated in Th1-associated Foxp3+ cells ( 1 . 68-fold ) . Strinkingly , Th1- Foxp3+ cells up-regulated a collection of transcriptional regulators , including Stat-3 ( 1 . 98 fold ) , Bcl6 ( 1 . 80-fold ) , Ap1 ( 2 . 14 fold ) and Runx2 ( 2 . 02 fold ) . Similarly , Th2-derived Foxp3+ cells co-expressed Blimp1 ( 3 . 78 fold ) , Tbx21 ( T-bet ) ( 2 . 64 ) , Hif2α ( 2 . 08 fold ) , E4bp4 ( 1 . 91 fold ) , Runx2 ( 1 . 68 fold ) and Egr2 ( 1 . 60 fold ) . These data suggest that there is either significant heterogeneity , or plasticity , within Foxp3+ populations [52] or that co-opting multiple transcription factors is common and does not restrict control to one particular T helper subset , but rather broadens regulatory function . Indeed , Treg cells isolated from Type-1 inflamed tissue had the capacity to suppress Th1 and Th2 cells , while Th2-Treg cells could only control Th2 cells . We hypothesize that suppression of Th2 cells by Th1-Treg cells could be mediated by TGF-β , which was slightly elevated in Th1- , but not Th2- , Treg cells ( Figure 4G ) and can potently inhibit Th2 cells [53] . However , given that TGFβ is highly regulated post-translationally , surface bound or secreted bioactive TGFβ may not be increased . Alternatively , the continued ability of Th1-Treg cells to control Th2 cells , but not Th1 cells , following over-expression of miR-10a , is most likely due to the increased IFNγ , which can also inhibit Th2 cell responses . Computational analysis [29] identified miR-182 in Th2-Foxp3+ cells and miR-10a in Th1-Foxp3+ cells as potential regulatory miRNA hubs , which targeted multiple differentially regulated genes . We focused on miR-182 and miR-10a for functional studies , as these were the top candidate regulatory hubs from the Monte Carlo analyses in Foxp3 cells from infected mice . In support of this , down regulated miR-10a and up-regulated miR-182 was also observed in Foxp3+ cells isolated from Th1- or Th2-inflammed lungs , analogous to the chronic infection studies . It was recently demonstrated that IL-2/STAT5 regulated miR-182 in helper and regulatory T cells [42] targeting Foxo1 and permitting helper cell proliferation . Despite the high consumption of IL-2 by Foxp3+ T cells and the requirement for Foxo1 , and Foxo3 , for Treg cell survival and function [54] , [55] , a role for miR-182 in Treg cells was not thoroughly investigated . Our systematic approach identified putative binding sites in the promoter of miR-182 for the IL-4-regulated transcription factor , cMaf . In agreement with this , IL-4-treated Treg cells up-regulated cMaf , similar to previous reports in macrophages and T cells [56] , [57] . Unlike naïve T cells , which produce IL-4 and IL-2 and up-regulate cMaf following IL-4 treatment , Treg cells did not produce IL-4 ( data not shown ) or IL-2 , in part through a miR-182-dependent pathway . The phosphorylation state of cMaf , additional pathways including IL-2 [42] and other transcriptional regulators may also contribute to miR-182 , as cMaf transcript levels in untreated iTreg and nTreg were indistinguishable from naive T cells , despite elevated miR-182 . Nevertheless , IL-4-treated Treg cells up-regulated cMaf and miR-182 , in line with other studies identifying that IL-4-treated human [58] , [59] and murine [60] Treg cells develop distinct and potent suppressive phenotypes . The precise mechanism from these studies , however , was unclear . It has long been appreciated that anergic and regulatory T cells do not produce IL-2 , through reduced JNK and ERK signaling [61] and remodeling of the Il2 locus [62] . We identified two miR-182-regulated genes that can control IL-2 production , Bach2 , a basic leucine zipper transcription factor [30] and Cd2ap [31] . As predicted , the up-regulation of cMaf and miR-182 by IL-4 led to a reduction of Bach2 and Cd2ap expression in Treg cells ( Figure S5 ) , with no IL-2 production . Disrupting this pathway , through inhibition of miR-182 , led to an increase in Bach2 and Cd2ap and a significant increase in transcription and secretion of IL-2 , indicating that IL-2 is critically regulated by miR-182 , potentially via control of Bach2 and Cd2ap . Although other important molecular pathways are under the control of miR-182 , including those controlled by C/EBPα , Arhgef3 and Hdac9 which are also intimately involved in Treg biology [63]–[66] , together with previous reports , we propose that IL-2 and IL-4 reinforce a negative feedback loop in Treg cells , with IL-2 induced [42] and IL-4-re-enforced miR-182 inhibiting IL-2 secretion . miR-10a was up-regulated in ex vivo Treg cells and naïve T cells polarized into iTreg with TGFβ in vitro [40] , [67] . We also observed an increase in miR-10a in ex vivo nTreg and iTreg cultures , relative to naïve T cells . However , our study design identified that miR-10a was subsequently reduced in Treg cells in Th1 environments . Whether splenic nTreg cells migrate to peripheral sites or de-novo iTreg cells respond to inflammatory events is unclear . To investigate the pathways involved in miR-10a regulation , we identified several putative TF binding sites in the miR-10a promoter , including the TGF-β [68] , IL-2 [69] , IL-12 [70] and IFNγ [71]-regulated transcription factor , CREB . CREB stabilizes Foxp3 in Treg cells [72] and is inhibited by IFNγ [71] , [73] , [74] . Creb expression was slightly elevated in ex vivo nTreg and in vitro-generated iTreg cells , relative to naïve T cells , but was successively decreased , below naïve T cell levels , following exposure to type-1 inflammatory signals , IL-12 and IFNγ . Furthermore , miR-10a followed a similar expression pattern as Creb , with reduced miR-10a following IL-12/IFNγ treatment , suggesting that Creb expression may influence miR-10a . Although multiple factors can influence miR-10a and Creb expression , these data indicate that Treg cells undergo dynamic molecular modifications upon exposure to various inflammatory signals , in this case along an IL-12/IFNγ , Creb , miR-10a axis . We identified several miR-10a-regulated genes in Foxp3+ cells , including Arrdc , an α-arrestin family member that degrades phosphorylated integrin β4 ( CD104 ) [75] and β2-adrenergic receptors [76] , two pathways required for the development [77] and survival [78] of Foxp3+ T cells . miR-10a also regulated the transcriptional repressor , Bcl6 , an important pathway recently identified in iTreg cells , preventing iTreg conversion in to TFH cells [40] . Furthermore , co-expression of Bcl6 with Blimp1 , Cxcr5 and PD-1 ( Pdcd1 ) in Foxp3+ in Treg cells identified as TFH-Reg cells , have also been reported [11] , [13] . Dissimilar to these studies we did not observe a TFH-Reg , or TFH phenotype , as phenotypic markers of TFH cells , beyond Bcl6 , were reduced or unchanged ( Cxcr5 −3 . 22-fold , Btla −2 . 0 fold , unchanged Il21 , Cd40l , Cd200 , Cd30l , Cd57 , and Fyn ) . The relatively subtle changes in miR-10a and Bcl6 in Th1-Treg cells may retain Treg function , without conversion into TFH cells , or TFH-Reg cells . For example , we observed that miR-10a was reduced 3 . 5-fold in Th1-Treg relative to naïve Treg cells , in contrast to the study identifying iTreg cell conversion into TFH cells [40] when iTreg cells were transduced with a miR-10a sponge to significantly sequester miR-10a . Similarly , we observed a relatively subtle increase in Bcl6 ( 1 . 79-fold , Table S1 ) compared to the ∼10-fold increase in TFH-Reg cells [11] , [13] . In addition to Bcl6 , we identified Fbxo30 ( also known as Fbxw7 and Fbw7 ) and the TGFβ-signaling molecule , Nr4a3 [79] , [80] as miR-10a-regulated genes in Th1-Treg cells . Conditional deletion of Fbxw7 in CD4+ cells [81] , or deletion of Nr4a3 and the closely related Nr4a1 , resulted in hyper-proliferation of T cells , thymic lymphoma's and lethal lymphoproliferation [82] , a phenotype similar to Foxp3−/− mice . Furthermore , ectopic expression of Nr4a3 induced Foxp3 expression and repressed IFNγ production [46] . IL-12/IFNγ treatment , which reduced Creb and miR-10a expression , resulted in a small increase in miR-10a-regulated genes , Fbxo30 and Nr4a3 and a small increase in Ifnγ transcription . Similar observations have been made in mouse and human Treg cells , with IL-12-treatment converting Foxp3+ cells into IFNγ+Foxp3+ cells [47] , [48] . Disrupting this molecular pathway , by over-expressing miR-10a , coupled with IL-12/IFNγ treatment , dramatically increased Ifnγ transcription , indicating that reduced miR-10a permitted tight control over IFNγ in Treg cells , possibly via Nr4a3 [46] . IFNγ secretion by Th1-Treg cells transfected with miR-10a mimics provides a plausible explanation as to how Th1-Treg cells retained their ability to partially control Th2 cells following miR-10a manipulation . Collectively , we have identified a suite of miR-10a targets in Th1-Foxp3+ cells , which regulate G-protein coupled receptor function ( Aardc3 ) , gene transcription ( Bcl6 ) , ion transport ( Clcn5 and Rap2a ) , iron metabolism ( Tfrc ) and TGF-β signaling ( Fbxo30/Fbxw7 and Nr4a3 ) . Furthermore , we have identified a mechanistic pathway of IL-12/IFNγ-regulated miR-10a expression that critically controls IFNγ production in Treg cells . In summary , Th1- or Th2-associated Foxp3+ cells developed distinct molecular profiles , influenced by local cytokine signaling pathways . IL-12/IFNγ-influenced miR-10a controlled subsequent IFNγ production in Th1-Treg cells , while IL-4-regulated miR-182 critically prevented IL-2 production in Th2-Treg cells . In addition , we propose that miR-182 and miR-10a function as regulatory hubs , coordinating a variety of pathways in Th2-Treg and Th1-Treg cells . These data strongly support the concept that different Foxp3+ cells activate distinct gene programs , shaped by different inflammatory signals . We also provide evidence for an upstream miRNA-mediated pathway regulating Foxp3+ cell specialization and functional stability .
Female C57BL/6 , C57BL/6 CD45 . 2 Foxp3gfp [83] , C57BL/6 Foxp3rfp [84] , C57BL/6 CD45 . 1 OTII RAG2−/− 6–8 weeks' old were bred and kept in the specific pathogen–free facility at the National Institute for Medical Research , or National Institutes of Health . All animal experiments were approved by UK National Institute for Medical Research Ethical Review Panel and NIAID animal care and use committee and carried out according to institutional guidelines ( UK National Institute for Medical Research Ethical Review Panel ) , UK Home Office regulations ( Project licence no . 80/2506 ) and according to The NIAID animal care and use committee in accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . A minimum of 5 mice per group was used in each experiment , unless indicated . Percutaneous infections were carried out with 35 S . mansoni cercariae ( Biomedical Research Institute , Rockville , MD ) , as previously described [85] . Mice were infected in the ear dermis with 105 L . major metacyclic promastigotes using a 27 . 5 G needle in a volume of 10 µl [38] . Cells were isolated from infected or inflamed tissue by mechanical disruption followed by percoll gradient separation and were stained with anti-mouse CD4 ( RM4-5 , BD Biosciences , Pacific Blue ( V450 ) or APC ) , CD3ε ( 17A2 , BD Biosciences , FITC or Alexa flour 700 ) , CD44 ( IM7 , BD Biosciences , PE-Cy7 or Alexa flour 700 ) , CD25 ( PC61 , BD Biosciences , PE or FITC ) and CD45 . 1 ( A20 , BD Biosciences , APC or PE ) diluted in PBS with 0 . 1% FCS before analysis using a BD LSRII and TreeStar FlowJo . For proliferation/suppression assays , 104 Teff cells were labeled with cell trace violet ( Invitrogen ) as per manufacturers guidelines and stimulated with irradiated splenocytes ( 2×105 ) and OVA ( 10 µg/ml ) for 3 days in the presence or absence of Treg cells , at the indicated ratios before analysis using a BD LSRII and TreeStar FlowJo . FACS purified cells were stored in RLT lysis buffer at −80°C until RNA was extracted . For mRNA analysis , RNA was extracted using RNeasy spin columns ( Qiagen ) followed by DNAse treatment . cDNA was generated from 5 ng of total RNA using WT-Ovation Pico system ( version 1 ) RNA Amplification System followed by double stranded cDNA synthesis using WT-Ovation Exon Module . cDNA quality was determined using an Agilent BioAnalyzer and through hybridization performance on Affymetrix GeneChip mouse gene 1 . 0 ST arrays . For miRNA analysis , small RNA species ( 20–200 bp ) were collected from the same samples and used for sequencing on the ABI SOLiD sequencer ( Applied Biosystems , Santa Clara , CA ) . Hybridization , fluidics and scanning were performed according to standard Affymetrix protocols ( http://www . affymetrix . com ) . GeneChip Operating Software ( GCOS v1 . 4 , http://www . Affymetrix . com ) was used to convert the image files to cell intensity data ( cel files ) . The array data were quantile normalized and analyzed using Partek Genomics Suite software ( Partek , inc . St . Louis , Mo . , v6 . 4-6 . 09 . 0129 ) . We identified differentially expressed genes using ANOVA and t-tests . Genes with false discovery rate corrected p-values less than 0 . 1 and fold change values ≥1 . 5 were considered significant . The resulting data were analyzed with IPA ( Ingenuity Pathway Systems , www . ingenuity . com ) . Libraries for SOLiD sequencing were prepared using the SOLiD Small RNA Expression Kit ( Applied Biosystems ) following the manufacturer's protocol . Templated beads for sequencing were prepared using a 1 pM library input following the Applied Biosystems SOLiD 3 Templated Bead Preparation Guide ( Applied Biosystems , Foster City CA ) . Small RNA libraries were run on the ABI SOLID 3 . 0 . Reads were mapped to Mus musculus microRNAs ( miRBase v13 . 0 ) [86] using the Small RNA Analysis Tool v0 . 4 ( Applied Biosystems ) . Read counts below 25 ( including miR-96 ) were removed from further analysis with read counts between samples normalized based on the total number of uniquely mapped reads in each sample . Candidate miRNA regulatory hubs were identified using Monte Carlo simulation analysis as described previously [29] . First , we used the seed-based target prediction algorithm TargetScanS to determine for each miRNA the number of predicted targets among our gene set of interest ( e . g . up/down-regulated transcripts in Foxp3+ cells in response to pathogen ) . We repeated this procedure 10 , 000 times with a new set of randomly selected genes from the genome each time , in order to generate a background expectation of the number of predicted target genes for each miRNA , which was then used to calculate an empirical p-value for the number of predicted target genes in the gene set of interest . To account for differences in the average 3′ UTR length between the genes of interest and the randomly selected genes in each simulation , the number of predicted target genes was normalized to the average 3′ UTR length . The genomic locations of the miR-182 and miR-10a transcription start sites ( TSS ) were identified using previously published methods [87] , [88] . We defined the promoter region as 1 kb upstream and 500 bp downstream of the TSSs . Within these promoters , we identified putative transcription factor binding sites using PWMSCAN [89] , which searches for sequences that match any known transcription factor binding site motif recorded in TRANSFACv10 . 2 . A match score with a p-value<5×10−6 was considered to be a high-confidence binding site prediction . RNA was isolated using RNeasy mini spin columns followed by miScript RT or Quantitect RT according to manufacturer's recommendations ( Qiagen ) . Real-time RT-PCR was performed on an ABI Prism 7900HT Sequence Detection System ( Applied Biosystems ) with relative quantities of mRNA determined using SYBR Green PCR Master Mix ( Applied Biosystems ) and by the comparative threshold cycle method as described by Applied Biosystems for the ABI Prism 7700/7900HT Sequence Detection Systems . mRNA levels were normalized to HPRT and miRNA levels were normalized to RNU6B and then expressed as a relative increase or decrease compared with levels in controls . Treg cells were isolated , as described above and transfected with 100 nM of miR-182 or miR-10a mimics or hairpin inhibitors ( Thermo Scientific Dharmacon ) or MOCK transfected using Nucelofection reagents according to manufacturer's recommendations ( Amaxa ) . Ex-vivo nTreg cells were cultured in rIL-2 ( 10 ng/ml ) -supplemented media for 24 hours before washing and use in suppression assays or transfer in-vivo . BlockiT fluorescent oligos ( Invitrogen ) were used to determine transfection efficiency . miRNA mediated impacts on mRNA expression was determined 24–48 hours post transfection . Naïve T cells ( CD4+CD44−CD62LhiCD25− ) were FACS purified and polarised under Th1 ( IL-12 , 10 ng/ml; anti-IL-4 , 10 ug/ml ) ) , Th2 ( IL-4 , 10 ng/ml; IL-2 , 10 ng/ml; anti-IFNγ , 10 ng/ml ) or iTreg ( TGFβ , 10 ng/ml , Retinoic acid , 10 nM ) conditions in the presence or absence of OVA-pulsed irradiated splenocytes as APC's for seven days , as indicated . Freshly isolated nTreg or in vitro generated iTreg cells were washed and cultured with either IL-4 ( 10 ng/ml ) , IL-12/IFNy ( both at 10 ng/ml ) or media only . Cells were harvested after 24 hours or supernatant was collected after 3 days . For adoptive transfer experiments , recipient mice were given OVA ( Sigma , Grade V ) via the trachea one day before adoptive transfer of 106 Th1 or Th2 cells . For intra-tracheal ( i . t . ) inoculation , mice were anaesthetized with ketamine and medetomidine and given 20 µl of OVA ( 10 µg ) in PBS directly into the trachea . Recipient mice were given OVA i . t . on day 1 and day 3-post transfer before analysis on day 4 . In some experiments , cells were isolated from recipient mice , transfected as above , and either adoptively transferred with newly generated Th1 or Th2 cells into a second recipient or used in proliferation/suppression . For 2nd adoptive transfer experiments , 106 newly generated Th1 or Th2 cells were co-transferred with 106 isolated and transfected Treg cells from recipient mice . Twenty-four hours after the OVA i . t . , mice were anaesthetized with pentobarbital . The trachea was cannulated and airspaces lavaged with 500 µl of sterile PBS for cellular analysis . For histopathological analysis lungs were removed , formalin ( 4% paraformaldehyde in PBS ) fixed embedded in paraffin and stained with Hematoxylin and eosin ( H&E ) . Inflammation was scored on an arbitrary 1–4+ basis taking into account both the degree of inflammation and its distribution . Local lymph nodes were isolated , prepared into a single cell suspension and cultured with OVA ( 10 µg/ml ) for 3 days . Cytokines were measured by ELISA using suppliers' guidelines . Capture and biotinylated detection antibodies for IL-4 , IL-5 , IL-10 , IFNγ , IL-17A and IL-9 were from R&D Systems . The concentration of analytes in the sample was determined from a serial-fold diluted standard curve with OD read at 405 nm in an ELISA reader . | The diversity of pathogens that the immune system encounters are controlled by a diverse suite of immunological effector responses . Preserving a well-controlled protective immune response is essential . Too vigorous an effector response can be as damaging as too little . Regulatory T cells ( Treg ) calibrate immune responses; however , how Treg cells adapt to control the diverse suite of effector responses is unclear . In this study we investigated the molecular identity of regulatory T cells that control distinct effector immune responses against two discrete pathogens , an intracellular parasitic protozoa , Leishmania major , and an extracellular helminth parasite , Schitsosoma mansoni . The two Treg populations studied were phenotypically and functionally different . We identified molecular pathways that influence this diversity and more specifically , we identified that two miRNAs ( miR-182 and miR-10a ) act as “regulatory hubs” critically controlling distinct properties within each Treg population . This is the first study identifying the upstream molecular pathways controlling Treg cell specialization and provides a new platform of Treg cell manipulation to fine-tune their function . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"parastic",
"protozoans",
"immunopathology",
"adaptive",
"immunity",
"immune",
"cells",
"leishmania",
"immunity",
"t",
"cells",
"immunology",
"protozoology",
"biology",
"microbiology",
"immunoregulation",
"host-pathogen",
"interaction",
"immune",
"response"
] | 2013 | miR-182 and miR-10a Are Key Regulators of Treg Specialisation and Stability during Schistosome and Leishmania-associated Inflammation |
Intestinal epithelial cells ( IECs ) compose the first barrier against microorganisms in the gastrointestinal tract . Although the NF-κB pathway in IECs was recently shown to be essential for epithelial integrity and intestinal immune homeostasis , the roles of other inflammatory signaling pathways in immune responses in IECs are still largely unknown . Here we show that p38α in IECs is critical for chemokine expression , subsequent immune cell recruitment into the intestinal mucosa , and clearance of the infected pathogen . Mice with p38α deletion in IECs suffer from a sustained bacterial burden after inoculation with Citrobacter rodentium . These animals are normal in epithelial integrity and immune cell function , but fail to recruit CD4+ T cells into colonic mucosal lesions . The expression of chemokines in IECs is impaired , which appears to be responsible for the impaired T cell recruitment . Thus , p38α in IECs contributes to the host immune responses against enteric bacteria by the recruitment of immune cells .
Attaching and effacing ( A/E ) bacterial pathogens , such as the enteropathogenic Escherichia coli ( EPEC ) and enterohemorrhagic E . coli ( EHEC ) , cause debilitating disease , especially among infants and children , and are a threat to global health [1] , [2] . Citrobacter rodentium ( C . rodentium ) is an A/E pathogen , which occurs naturally in mice , and serves as an excellent animal model for these mucosal infections [3] , [4] . C . rodentium has a remarkable ability to colonize the murine colon and cecum , but is typically subclinical and self-limiting , and is eventually cleared from the gastrointestinal tracts in immunocompetent mice [3] . Studies of C . rodentium infection in immunodeficient mice have established that CD4+ T cells and C . rodentium-specific antibody responses are essential components of adaptive immunity for eradicating the infection [5] , [6] , and recent studies have revealed that TH1 and TH17 immune responses have important host defense functions during C . rodentium infection [7]–[9] . However , the molecular mechanism by which these immune responses are regulated after the mucosal surface of the intestinal tract is stimulated by pathogens is still largely unknown . The role of the NF-κB pathway in intestinal epithelial cells was reported recently using IKK subunit knockout mice [10] , [11] . The NF-κB pathway in intestinal epithelial cells is essential for intestinal immune homeostasis , although the mechanisms are not exactly the same , as one study reported dysregulated epithelial cell integrity while another reported dysregulated immune cell function after different pathogen infections [10] , [11] . These results tempted us to explore the role of p38α , another major inflammatory pathway , in intestinal epithelial cells and its role in immunity to enteric pathogens . p38α is the prototypic member of the p38 group of mitogen-activated protein kinases ( MAPKs ) [12] , and its activation has a pivotal role in linking inflammatory stimuli to cellular responses [13]–[15] . Previous studies using a human colon epithelial cell line ( Caco-2 ) have shown a role for p38α in enteric pathogen-induced IL-8 production [16] , but the role of p38α in intestinal epithelial cells in vivo is not known . The embryonic lethality of p38α-null mice and the limited target specificity of p38 inhibitors on p38α are limiting factors for understanding the role of p38α in vivo . Here we used C . rodentium infection and mice lacking p38α in intestinal epithelial cells to study the role of p38α in host responses to mucosal infection . We found that unlike the NF-κB pathway , which controls intestinal immune homeostasis , intestinal epithelial p38α is crucial for immune cell recruitment in the colonic mucosa . The different inflammatory signaling pathways appear to differentially affect immune responses in intestinal epithelial cells .
C . rodentium is a popular surrogate mouse model for the study of attaching and effacing bacterial pathogens . Their attachment to mouse colonic epithelial cells results in effacement of the brush border , termed an A/E lesion , and colonic mucosal hyperplasia [17] . To investigate the function of p38α in the intestinal epithelium , we generated mice lacking p38α in intestinal epithelial cells ( VillinCre-p38ΔIEC ) by crossing loxP-flanked ( p38αfl/fl ) mice with villin-Cre ( VillinCre ) -expressing mice . These mice appear healthy and have no remarkable histological abnormalities in the intestine , currently studied up to seven months after birth . Lack of p38α protein in the intestinal epithelial cells from VillinCre-p38ΔIEC mice was confirmed by immunoblotting ( Fig . 1A ) . C . rodentium infection induced p38α phosphorylation in the intestinal epithelial cells of p38αfl/fl mice ( Fig . 1A ) , indicating an involvement of p38α in the C . rodentium-induced host response . C . rodentium inoculation induced rapid and transient body weight loss in both p38αfl/fl and VillinCre-p38ΔIEC mice; however , VillinCre-p38ΔIEC showed impaired body weight recovery after 7 days of infection ( Supplementary Fig . S1 ) . The difference between wildtype and VillinCre-p38ΔIEC mice was moderate but statistically significant ( Supplementary Fig . S1 ) . We further analyzed bacterial burden in the colon tissues of p38αfl/fl and VillinCre-p38ΔIEC mice and found it to be comparable at the early times of infection , but much worse in VillinCre-p38ΔIEC mice after two weeks of infection ( Fig . 1B and Supplementary Fig . S2 ) . Moreover , the eventual clearance of the bacteria occurred later in VillinCre-p38ΔIEC mice ( Fig . 2B ) , indicating that VillinCre-p38ΔIEC mice exhibit a significant defect in clearing bacteria from the colon tissues . Immunohistological studies showed that at 1 week after infection , C . rodentium localized close to the surface of the colon epithelial cells similarly in p38αfl/fl and VillinCre-p38ΔIEC mice ( Fig . 1C ) . However , at two weeks after infection , p38αfl/fl mice showed only a slight bacterial staining on the colon surfaces , whereas numerous C . rodentium still remained in VillinCre-p38ΔIEC mice ( Fig . 1C ) . The greater bacterial burden recovered from the colons of VillinCre-p38ΔIEC mice two-weeks after infection was confirmed by qPCR to quantify bacterial 16s rDNA ( Supplementary Table S1 ) . H&E staining using adjacent sections showed inflammatory cell invasion into the colonic mucosa at two weeks after infection ( Fig . 1D ) . However , the degree of inflammatory cell infiltration was more severe in p38αfl/fl mice two weeks after infection ( Fig . 1D and 1E ) , but the bacterial burden was less in those mice compared with the VillinCre-p38ΔIEC mice ( Fig . 1B and 1C ) . These results indicate that p38α in intestinal epithelial cells is involved in the clearance of infected C . rodentium , and the role of epithelial p38α in inflammatory cell invasion is at least part of the underlying mechanism . NF-κB signaling , a well-known major inflammatory pathway , has been explored in the gut recently [10] , [11] . Deletion of IκB kinase-β ( IKKβ ) or IKKγ in intestinal epithelial cells causes abnormal epithelial integrity and subsequent abnormal spontaneous inflammation [11] or impaired conditioning of dendritic cells and subsequent impaired T cell polarization after parasite inoculation [10] . Here we examined the epithelial integrity and immune cell functions in our VillinCre-p38ΔIEC mice . TdT-mediated dUTP nick end labeling ( TUNEL ) staining of colon tissues before and after C . rodentium infection revealed no significant differences in epithelial cell viability between p38αfl/fl and VillinCre-p38ΔIEC mice ( Fig . 2A and data not shown ) , although more intestinal epithelial cells , especially located at the bottom of the mucosa , showed evidence of apoptosis after C . rodentium infection in both animals ( Fig . 2A ) . Expression of claudin-1 and claudin-2 , members of the tight junction protein family that regulate epithelial permeability , were analyzed by qRT-PCR using colon tissues from C . rodentium-infected and uninfected mice . As reported [18] , [19] , claudin-2 was strongly induced while claudin-1 expression was not affected by C . rodentium-infection ( Fig . 2B ) . Deletion of p38α in intestinal epithelial cells did not affect claudin-1 and claudin-2 expression as compared to infected and uninfected p38αfl/fl mice ( Fig . 2B ) . This data supports the conclusion that intestinal epithelial integrity is not affected by p38α deletion . Consistently , we did not detect bacterial translocation into the colon mucosa in either p38αfl/fl or VillinCre-p38ΔIEC mice by IHC ( Figure 1C and Supplementary Fig . S3 , the adjacent sections were used in Figure 1C , S3 , 2A , and 1D ) and by qRT-PCR to quantify bacterial 16s rDNA ( data not shown ) . We then analyzed immune cells in the mesenteric lymph nodes , as these nodes drain the murine large bowel , potentially in concert with the caudal lymph nodes . The size of mesenteric lymph nodes was similar in p38αfl/fl and VillinCre-p38ΔIEC mice . To examine dendritic cells ( DC ) , the composition and function of DC subsets in the intestine-associated lymphoid tissue of p38αfl/fl and VillinCre-p38ΔIEC mice was examined after C . rodentium infection . Similar frequencies of CD11c+CD11b−CD8α− ( double-negative , DN ) , CD11c+CD11b−CD8α+ ( CD8α+ ) , and CD11c+CD11b+CD8α− ( CD11b+ ) DC subsets were observed in mesenteric lymph node cells of p38αfl/fl and VillinCre-p38ΔIEC mice at one and two weeks after C . rodentium infection ( Fig . 2C ) . No significant difference in tumor necrosis factor ( TNF ) - α production by the CD11c+CD11b+CD8α− ( CD11b+ ) , CD11c+CD11b−CD8α+ ( CD8α+ ) , or CD11c+CD11b−CD8α− ( double negative ) DC subset population of the draining mesenteric lymph nodes was observed ( Supplementary Fig . S4A , S4B , S4C ) . In addition , no T cell functional polarization shift was observed , as similar amounts of IFN-γ ( TH1 response ) and IL-17 ( TH17 response ) were produced after bacterial antigen stimulation of draining mesenteric lymph node cells of p38αfl/fl and VillinCre-p38ΔIEC mice at 1 and 2 weeks after C . rodentium infection ( Fig . 2D , 2E ) . In supporting this notion , the CD4+T cells from mesenteric lymph node or lamina propria of C . rodentium-infected p38αfl/fl and VillinCre-p38ΔIEC mice showed similar expression of IFN-γ and IL-17 ( Supplementary Fig . S5 and S6 ) . In addition , the production of TNF by macrophages in draining mesenteric lymph nodes was also comparable in the p38αfl/fl and VillinCre-p38ΔIEC mice ( Fig . 2F ) . These results indicate that , unlike the ablation of the NF-κB pathway in intestinal epithelial cells , epithelial integrity and immune cell in the intestine-associated lymphoid tissue are normal in VillinCre-p38ΔIEC mice . Although the functions of immune cells isolated from mesenteric lymph nodes and lamina propria of VillinCre-p38ΔIEC mice appeared to be normal when compared with that of p38αfl/fl mice , the expression of various TH1 and TH17 related cytokines in the colon , including IFN-γ , IL-17 , and IL-22 , which are critical factors in the host defense against A/E bacterial pathogens [9] , [20] , was significantly less in VillinCre-p38ΔIEC mice two weeks after C . rodentium infection ( Fig . 3A , B , C ) . Similar expression patterns of KC and IL-6 , but not TNF , were also observed ( Fig . 3D , E , F ) . Immune cells in colon include resident cells in lamina propria and infiltrated cells in the colonic mucosa . Because there is no functional difference between the immune cells isolated from lamina propria of VillinCre-p38ΔIEC and p38αfl/fl mice ( Supplementary Fig . S6 ) , the differences shown in Fig . 3E are likely caused by infiltrated immune cells . Therefore , we determined whether deletion of p38α in IEC affects the number of infiltrated TH17 cells in colon . Immunostaining showed that there were more infiltrated TH17 cells in the colons of p38αfl/fl mice in comparison with that of VillinCre-p38ΔIEC mice , and the majority of TH17 cells were CD4+ cells ( Fig . 3G ) . Flow cytometry analysis of cells isolated from colon tissues confirmed this result ( Fig . 3H ) . Since the degree of inflammatory cell infiltration was less severe in VillinCre-p38ΔIEC mice in comparison with p38αfl/fl mice ( Fig . 1D , 1E and 4A ) , the infiltration of immune cells into the colonic mucosa was examined in more detail . Various frozen sections of colon were stained with antibodies against CD4 , CD11c ( as a DC marker ) , Gr-1 ( as a neutrophil marker ) , and F4/80 ( as a macrophage marker , data not shown ) . No staining of any marker used here was detected in the colon tissues of either p38αfl/fl and VillinCre-p38ΔIEC mice before C . rodentium infection ( Fig . 4B and data not shown ) , and a few scattered immune cells that had infiltrated into the mucosa were detected at 1 week after infection ( Supplementary Fig . S7 ) . Infiltration of immune cells , especially CD4+ T cells , was dramatically increased in p38αfl/fl mice two weeks after infection ( Fig . 4C ) , consistent with previous reports that CD4+ T cells infiltrate into the colonic mucosa and play a central role in the clearance of this bacterium [5] . In contrast , the infiltration of CD4+ T cells into the colonic mucosa of VillinCre-p38ΔIEC mice two weeks after C . rodentium infection was much less than that in p38αfl/fl mice ( Fig . 4C , 4D and Supplementary Fig . S8 ) . FACS analysis of CD4+T cells in isolated lamina propria cells confirmed the decreased CD4+T cell infiltration into the colonic mucosa in VillinCre-p38ΔIEC mice ( Fig . 4E ) . These results suggest that p38α in intestinal epithelial cells is required for the CD4+ T cell recruitment into the colonic mucosa . Because p38α is important for cytokine expression [13]–[15] , the expression of chemokines in the colonic epithelial cells of p38αfl/fl and VillinCre-p38ΔIEC mice one week after C . rodentium infection was determined by microarray analysis ( Fig . 5A , B , and C ) . The time point was chosen as the time when the bacterial burden is still similar between p38αfl/fl and VillinCre-p38ΔIEC mice and when immune cells start to infiltrate into the colonic mucosa as described above . Although Cmtm6 was upregulated similarly in both p38αfl/fl and in VillinCre-p38ΔIEC mice after infection , the expression of most genes in p38αfl/fl mice colon epithelial cells that were upregulated by infection did not change in VillinCre-p38ΔIEC mice colon epithelial cells ( Fig . 5A , B and C ) . Rather , the expression of some genes was downregulated , even after infection , compared with those in uninfected p38αfl/fl control mice colon epithelial cells ( Fig . 5B ) . We confirmed the differential expression of the selected genes by qRT-PCR ( Fig . 5D ) , finding that p38α deletion indeed reduced the expression of a number of C . rodentium infection-induced chemokines in colon epithelial cells . We further studied two chemokines , Ccl25 ( also known as TECK ) and Cxcl10 ( also known as IP-10 ) , to evaluate whether their associated impairment in expression following p38α deletion might contribute to the phenotype of the VillinCre-p38ΔIEC mice . Analyzing C . rodentium infection in Caco-2 cells revealed that the expression of Ccl25 and Cxcl10 was directly induced by C . rodentium in a p38-dependent manner ( Supplementary Fig . S9 ) . Immunofluorescence revealed lower induction of Ccl25 in the colonic epithelial cells of VillinCre-p38ΔIEC mice after C . rodentium infection in comparison with p38αfl/fl mice ( Fig . 6A ) . Since Ccl25 is one of the CD4+ T cell-recruiting molecules [21] , the loss of Ccl25 induction should contribute to the impaired recruitment of CD4+ T cells into the colonic mucosa of VillinCre-p38ΔIEC mice . Cxcl10 is a chemoattractant for monocytes/macrophages , T cells , NK cells , and dendritic cells , and it promotes T cell adhesion to endothelial cells . Since mice lacking this gene are available , we assessed its role in host defense . Analysis of C . rodentium-infection in Cxcl10 knockout mice revealed that bacterial clearance was impaired in C . rodentium-infected Cxcl10 knockout mice ( Fig . 6B ) . As observed in VillinCre-p38ΔIEC mice , the induction of IL-17 and IFN-γ was impaired in Cxcl10 knockout mice ( Fig . 6C ) . Unlike VillinCre-p38ΔIEC mice , the induction of KC and IL-6 was not affected , but TNF induction was inhibited by Cxcl10 deletion ( Fig . 6C ) . Despite the similarities in phenotype between VillinCre-p38ΔIEC and Cxcl10 mice in response to C . rodentium , the differences were also anticipated , since Cxcl10 induction should be only part of the mechanism of p38α-mediated host-defense against C . rodentium infection . Parasite-induced RELM-β and Gob5 in intestinal epithelial cells can be blocked by deletion of IKK-β[10] , [11] , whereas their expression was not affected by deletion of p38α ( Fig . 6D and data not shown ) . In addition , expression of S100A8 , an antibacterial protein , was lower in VillinCre-p38ΔIEC mice , while other antibacterial peptides , Defensin-α and RegIIIβ , were similarly expressed ( Fig . 6D ) . These data again show that intestinal epithelial p38α and NF-κB have different functions in initiating immune responses from the mucosal surface of the intestinal tract , and p38α is required for chemokine expression in the intestinal epithelial cells , which have crucial roles in recruiting immune cells into the colon mucosa upon bacterial infection .
Recent findings have shown that blocking NF-κB signaling in intestinal epithelial cells leads to dramatic impairments in mucosal immune responses and/or dysregulated intestinal epithelial cell integrity [10] , [11] . Because MAP kinases represent another major inflammatory pathway in the gut that has not been explored in detail , we addressed the role of p38α in intestinal epithelial cells in the context of an A/E bacterial infection using mice with p38α-specific deletion in intestinal epithelial cells . Here we reported that , unlike NF-κB pathway deletion , p38α deletion is not involved in the dysregulation of immune cell function or epithelial cell integrity , but it is involved in the dysregulated chemokine expression and subsequent immune cell recruitment to the infected lesions . While the host immune response against C . rodentium is still not fully characterized , it is known to involve TH1 and/or TH17 cells [7] , [9] . The importance of CD4+ T cells in this infection has been demonstrated by the fact that C . rodentium infection is fatal in mice lacking CD4+ T cells [5] . While we do not fully understand how localized CD4+ T cells recruited to the infected lesion are involved in fighting this bacterial infection , previous studies have implicated T cells in much of the tissue pathology seen during infection , including mucosal hyperplasia [6] , and inflammation that may limit C . rodentium survival/colonization at the mucosal surface . Similarly , CD4+ T cell dependent IgG antibodies are required for survival and clearance of this pathogen [5] . Therefore , the observed defect in CD4+ T cell recruitment to the intestinal mucosa in VillinCre-p38ΔIEC mice is likely to be the cause of their impaired defense against C . rodentium . This defect should also impair the “amplification cycles” of the immune response in the infected lesions , since the reduced immune cell recruitment is linked to the subsequent attenuation in cytokine production from epithelial cells . In fact , the reduced expression of S100A8 , which is one of the antimicrobial proteins against C . rodentium and is regulated by IL-22 [20] , was also detected in VillinCre-p38ΔIEC epithelial cells after infection . It should be noted that many bacterial pathogens aside from C . rodentium induce p38α phosphorylation in intestinal epithelial cells , usually as an innate driven response to their products such as flagellin [16] . Thus it will be interesting to see whether the critical role played by p38 signaling in the host response to C . rodentium can be replicated in other infection models . Moreover the importance of p38α in the host response to A/E pathogens is further highlighted by the fact that these pathogens are known to suppress p38 activation in infected IEC , through the actions of their type III secretion systems [16] . While the mechanisms and bacterial effector proteins involved in this subversion have yet to be identified , our current studies suggest these actions may prove to be protective to the pathogen by limiting the recruitment of T cells to the gut . Although we show here that p38α deletion in intestinal epithelial cells is linked with reduced chemokine expression and reduced immune cell recruitment to the lesions , we cannot fully exclude the possibility that for some pathogens , p38α is also related to the impairment of other immune functions in intestinal epithelial cells , as differential blocking of NF-κB pathway components during infection by different pathogens induced different host reactions . Nonetheless , our study defines a crucial role for p38α in intestinal epithelial cells for triggering the host immune responses in the gastrointestinal tract . The different function of different signaling pathways must be taken into consideration in the development and application of anti-inflammatory agents .
2×109 CFU C . rodentium strain DBS 100 ( ATCC 51459; American Type Culture Collection , Manassas , VA ) in a total volume of 200 µl was orally inoculated into each mouse after fasting for 8 hours . The concentration of bacteria was measured by absorbance at optical density 600 , and was serially diluted and seeded on a MacConkey agar ( Difco Laboratories , Sparks , MD ) plate to confirm the CFU administered . Body weight changes were monitored daily . Citrobacter antigen was prepared as previously described [22] . Briefly , C . rodentium culture was washed with ice-cold PBS and sonicated on ice . The homogenate was then centrifuged at 4°C for 30 min . Supernatants were collected and sterilized by 0 . 22 µm filtration , and protein concentrations were determined . After euthanizing mice , entire colon and mesenteric lymph nodes were removed under aseptic conditions . The terminal 0 . 5-cm piece of the colon was weighed , homogenized , serially diluted , and plated in triplicate on MacConkey agar plates to quantify bacterial numbers . To measure the bacterial 16s rDNA , tissue DNA was prepared from the colon of the infected mice using DNeasy Blood & Tissue Kit ( Qiagen , Valencia , CA ) . Quantitative real-time PCR was performed using 50 ng of DNA and bacterial universal primers for r16 ( forward; TCCTACGGGAGGCAGCAGT , and reverse; GGACTACCAGGGTATCTAATCCTGTT ) . GAPDH level was measured as a reference . The adjacent 0 . 5-cm piece was fixed in 10% formalin for H&E and C . rodentium staining , or frozen in optimal cutting temperature media ( Tissue-Tek , Elkhart , In ) for staining of other cell markers and Ccl25 expression . Immunostaining was performed as described previously [23] . Rabbit anti-Citrobacter antibodies ( kindly provided by Dr . David B . Schauer; MIT ) were used to identify adherent C . rodentium [3] , and Rabbit anti-mouse Ccl25 antibodies were purchased from Santa Cruz Biotechnology ( Santa Cruz , CA ) to identify Ccl25 expression . Alexa Fluor 594 conjugated anti-rabbit IgG ( Molecular Probes ) was used for visualization . Alexa 488 conjugated CD4 ( GK1 . 5 ) , Gr-1 ( RB6-8C5 ) , and CD11c ( N418 ) antibodies were purchased from BioLegend ( San Diego , CA ) . Unconjugated anti-CD4 ( GK1 . 5 ) and anti-IL-17 antibodies were obtained from Santa Cruz Biotechnology ( Santa Cruz , CA ) . Slides were mounted using VectaShield with DAPI ( Vector Labs , Burlingame , CA ) . The in situ Cell Death Detection Kit ( Roche , Mannheim , Germany ) was used for TUNEL staining ( TdT-mediated dUTP nick end-labeling ) . The degree of inflammatory cell infiltration was assessed by a histological score . The score was defined as a scale of 0–3 as follows: inflammatory cell infiltration; 0 = occasional inflammatory cells in the lamina propria; 1 = increased number of inflammatory cells in the lamina propria; 2 = confluent inflammatory cells , extending into the submucosa; 3 = transmural extension of the infiltrate . Colon epithelial cells were isolated using a modified rapid low-temperature method as described previously [24] . Briefly , the entire colon was removed and washed with ice-cold PBS . After dividing the intestine into 2–3 mm long fragments and transferring them into chelating buffer ( 27 mM trisodium citrate , 5 mM Na2PO4 , 96 mM NaCl , 8 mM KH2PO4 , 1 . 5 mM KCl , 0 . 5 mM DTT , 55 mM D-sorbitol , 44 mM Sucrose , 6 mM EDTA , 5 mM EGTA , pH 7 . 3 ) for 45 min . at 4°C , epithelial cells were then dissociated by repeated vigorous shaking . Tissue debris was removed by a cell-strainer ( 100 µm ) and colon epithelial cells were collected by centrifugation at 150×g for 10 min . at 4°C . The viability of colon epithelial cells was confirmed by trypan blue staining and processed for protein or RNA extraction . Total cell extracts from colonic epithelial cells were analyzed by SDS-polyacrylamide gel electrophoresis and transferred to polyvinylidene difluoride membranes ( Hybond-P; Amersham Pharmacia Biotech , Buckinghamshire , UK ) , followed by immunoblotting with anti-p38α , anti-phosphorylated p38 ( Cell Signaling Technology , Danvers , MA ) , and anti-GAPDH ( Chemicon , Temecula , CA ) antibodies . The bound antigens were detected using SuperSignal West Femto Maximum Sensitivity Substrate ( Pierce , Rockford , IL ) . Mesenteric lymph nodes were aseptically prepared and dendritic cells were isolated by positive selection using CD11c+ MACS microbeads ( Miltenyi Biotec ) . Cells were stained with anti-CD11c-APC , CD11b-PerCP , and CD8α-FITC antibodies ( eBioscience ) . Intracellular TNF staining was performed using Fixation/Permeabilization buffers and anti-TNF-PE antibodies ( eBioscience ) . Stained samples were analyzed using a FACScalibur flow cytometer and FlowJo software . Lymphpcytes isolated from mesenteric lymph nodes were cultured at a concentration of 5×106 cells/ml and restimulated with 50 µg/ml of Citrobacter antigen for 48 hours . Culture supernatants were prepared to measure the IL-17 and IFN-γ concentrations by ELISA ( R&D systems ) . The colon was removed and opened longitudinally , then washed with ice-cold PBS to remove debris . The tissue was then cut into small pieces ( ∼1 cm ) and further incubated for 30 min . at 37°C with gentle shaking in HBSS with 1 mM DTT and 2% FCS , and the supernatant was removed . The colon tissue was further incubated in HBSS with 1 mM EDTA and 2% FCS for 30 min . at 37°C with gentle shaking . Tissue was collected and further cut into smaller pieces , and digested with 0 . 5 mg/ml collagenase type IV ( Sigma-Aldrich . St . Louis , MO ) at 37°C with gentle shaking for 2 hrs . Cells were washed in HBSS twice and passed through a 40 µm cell strainer . Whole colon cells were resuspended in RPMI-1640 medium supplemented with 10% FBS and antibiotics , and treated with PMA and ionomycin for 6 hours . Intracellular staining of cytokines was performed using Cytofix/Cytoperm Fixation/Permeabilization Solution kit ( BD Bioscience , San Jose , CA ) . Cells were harvested and stained with anti-CD4-PE and anti-IL-17-APC antibodies to measure the infiltration of CD4 cells and the expression of IL-17 in the whole colon . Lamina propria cells were harvested by discontinuous 40%/80% Percoll gradient centrifugation of whole colon cells . After centrifugation , cells in the interface were collected and washed twice in HBSS . After 6 hours of PMA and ionomycin treatment , cells were harvested and stained with anti-CD3-FITC , anti-CD4-PE , anti-IFN-γ-PerCP , and anti-IL-17-APC antibodies for flow cytometry analysis . Entire colons were removed and cultured at 37°C for 24 hours as described previously [20] . Supernatants were collected and IL-17 , IFN-γ , IL-22 , KC , IL-6 , and TNF levels were analyzed by ELISA ( R&D systems ) . Caco-2 human colonic epithelial cell lines ( ATCC ) were grown in DMEM supplemented with 10% FBS without antibiotics . Seven days after reaching confluency , the cells were infected with C . rodentium at a multiplicity of infection of 50 . After 4 hours of incubation , as described previously [16] , the cells were washed and RNA was extracted to assay the cell responses . In the case of using a p38 inhibitor , SB203580 ( Calbiochem , San Diego , CA ) was added at 5 nM for 1 hour prior to infection . Total RNA from isolated colonic epithelial cells and Caco-2 cells was isolated using Trizol reagent ( Invitrogen , Carlsbad , CA ) and analyzed by chemokine & receptor oligomicroarrays ( Oligo GEArray OMM-022 or OHS-022; SABiosciences , Frederick , MD ) according to the manufacturers' instructions . Colon tissues from Citrobacter-infected or uninfected mice were obtained and total RNA was prepared and cDNA was synthesized by reverse transcription . Microarray data analyses were performed using GEArray Expression Analysis Suite version 2 . 0 software , according to the manufacturer's instructions ( SABiosciences ) . The expression threshold was determined to be when the average density of the spot is more than the mean value of the local backgrounds of the lower 75th percentile of all spots . Quantitative real-time PCR was performed using a TaqMan gene expression system with Sybr Green ( Applied Biosystems , Foster City , CA ) . The primer sequences are listed in Table S2 . All values were normalized to the level of the house keeping gene GAPDH messenger RNA , and relative expression was calculated according to the ΔΔCT method . The statistical significance of the differences between the two groups was determined using the Student's t test when variances were equal , or using the Welch's t test when variances were unequal . | The cellular responses of intestinal epithelial cells ( IECs ) to microorganisms in the gastrointestinal tract are mediated by activation of a number of intracellular signaling pathways . It was shown that the NF-κB pathway in IECs is essential for epithelial integrity and intestinal immune homeostasis , and here we show that p38α-mediated signaling in IECs is not important for epithelial integrity and immune cell function , but is critical for the clearance of the infected pathogen . p38α in IECs is essential for pathogen-induced chemokine expression in IECs and for subsequent immune cell recruitment into the intestinal mucosa , which leads to the clearance of the infectious pathogen . Our results indicate that different intracellular signaling pathways in IECs mediate distinct cellular responses to microorganisms in the gastrointestinal tract , and this information should be taken into consideration in the development of pathway-targeted therapeutic interventions for gastrointestinal infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"gastroenterology",
"and",
"hepatology/gastrointestinal",
"infections",
"immunology/immune",
"response",
"infectious",
"diseases/bacterial",
"infections",
"immunology/immunity",
"to",
"infections",
"infectious",
"diseases/gastrointestinal",
"infections"
] | 2010 | Epithelial p38α Controls Immune Cell Recruitment in the Colonic Mucosa |
Sensory processing is associated with gamma frequency oscillations ( 30–80 Hz ) in sensory cortices . This raises the question whether gamma oscillations can be directly involved in the representation of time-varying stimuli , including stimuli whose time scale is longer than a gamma cycle . We are interested in the ability of the system to reliably distinguish different stimuli while being robust to stimulus variations such as uniform time-warp . We address this issue with a dynamical model of spiking neurons and study the response to an asymmetric sawtooth input current over a range of shape parameters . These parameters describe how fast the input current rises and falls in time . Our network consists of inhibitory and excitatory populations that are sufficient for generating oscillations in the gamma range . The oscillations period is about one-third of the stimulus duration . Embedded in this network is a subpopulation of excitatory cells that respond to the sawtooth stimulus and a subpopulation of cells that respond to an onset cue . The intrinsic gamma oscillations generate a temporally sparse code for the external stimuli . In this code , an excitatory cell may fire a single spike during a gamma cycle , depending on its tuning properties and on the temporal structure of the specific input; the identity of the stimulus is coded by the list of excitatory cells that fire during each cycle . We quantify the properties of this representation in a series of simulations and show that the sparseness of the code makes it robust to uniform warping of the time scale . We find that resetting of the oscillation phase at stimulus onset is important for a reliable representation of the stimulus and that there is a tradeoff between the resolution of the neural representation of the stimulus and robustness to time-warp .
In recent years , there has been a growing interest in understanding how temporal information of sensory stimuli is encoded by sensory corticies ( see , e . g . , [1]–[8] ) . It has been shown that information about the features of the external stimulus is encoded in the fine temporal structure of the neural response ( see , e . g . , [8]–[15] ) . We are especially interested here in stimuli that have a natural hierarchy of temporal scales , such as speech and its components , including phones , diphones , words etc . Sensory processing has also been shown to be associated with the appearance of gamma oscillations in various sensory corticies ( see , e . g . , [16]–[20] ) . This raises the question whether the gamma oscillations can be directly involved in the representation of time-varying stimuli , including stimuli whose time scale is larger than that of a gamma cycle . Such a model was suggested by Hopfield [5] , and later was studied in the contex of diphone discrimination [21] . In this model subthreshold oscillatory input acts to coordinate the firing of cells so that a downstream neuron can read out a population code based on synchrony of firing . The implementation of this idea had a memory of about 200 ms , in a way that varied along a given stream of speech; the time scale of the memory depended on a dynamically changing “Lyapunov exponent”; the more negative this quantity , the shorter the memory and the more stable the representation . Thus , the longer memory was also associated with a less stable and less transparent representation . Here we build on the ideas in that paper about the synchronizing effects of gamma oscillations . However , to represent a signal having a natural time scale of more than one gamma period , we use multiple periods explicitly in the representation . The aim of this paper is to show that this idea can be implemented robustly in the context of biophysically reasonable networks of neurons . The gamma oscillations are a product of the network , rather than an external input , and correspond to spiking events in the network , not subthreshold oscillations . We use a dynamical model of a network of spiking cells [22] that responds to a one-dimensional time-varying input in the shape of a sawtooth . Such a signal models the response of one cochlear frequency-band to a short speech stimulus , such as a diphone , that lasts several gamma cycles . We show that the oscillations produced by the network tend to discretize the neural response to the sawtooth . From this , we get a binary response of the population , based on which cells fire in which cycles . Using a simple measure of discriminability , we examine the reliability of the representation , and show that reliability requires an onset signal , something that is well known for sensory signals ( see , e . g . , [14] , [23] , [24] , [25] ) . We also show that the representation is robust to moderate noise and time warp . In the Discussion , we compare the ideas of this paper with other work on coding ( or recognition ) of temporal patterns . We also discuss how hierarchies of oscillations in the nervous system may relate to the natural hierarchy of timescales in speech ( phone , diphone , syllable , word , and sentence ) and possible mechanisms for reading out the kind of code we suggest . Ultimately , we would like to study the representation of a diphone . A diphone is a speech segment , roughly from the middle of a phoneme to the middle of the phoneme following it . In a single cochlear frequency-band , the temporal fluctuations of the sound energy of a diphone can be represented in caricature by a single sawtooth waveform that mimics the dynamics of energy as it enters and leaves the frequency band . In this study we focus on the representation of sawtooth-shaped signals . Different sawtooths will be represented by a single shape parameter , , that specifies the time of the energy peak in the sawtooth from the beginning of the sawtooth , in units of the sawtooth period ( see Figure 1 ) . Unless otherwise stated we use a typical duration of 50 ms for the sawtooth stimulus , although we have tested the network response for slightly shorter and longer stimulus durations 40–100 ms . The advantage of using a simplistic abstract model for the input stimulus , instead of , for example , a real intensity profile taken from speech , is that it allows for systematic investigation of the representation which , in turn , facilitates the clear understanding of the properties of the representation . The functional architecture of the network is depicted in Figure 1 . The excitatory-inhibitory interactions are sufficient to generate and sustain oscillations in the gamma frequency range . Specifically , oscillation period was about 18 ms . Hence , the duration of the external stimulus ( typically 50 ms ) is about three network cycles . The oscillations are generated via a mechanism known as PING ( Pyramidal-Interneuronal Network Gamma ) . Essentially , input from the excitatory cells cause the inhibitory population to fire and generate a volley of inhibition that synchronizes the network activity ( see [22] for a fuller description ) . Excitatory cells are further divided into three functional subpopulations according to their different inputs . The background subpopulation receives high DC current and is responsible for generating the intrinsic gamma oscillations . The onset subpopulation receives an onset signal and is responsible for resetting the oscillation phase to synchronize it with the stimulus onset . The last subpopulation is the coding population that receives the time dependent sawtooth input current . A more detailed description of the network and its dynamics appears in the Materials and Methods section below .
Figure 2 shows three examples of the population response to the external stimuli , in the absence of internal noise . The x-axis is time and every line shows the spiking events of a different cell in the population during the same trial . The cells are ordered according to their functional subpopulation . At the bottom ( cells 1–30 ) is the excitatory background population that , together with the inhibitory population ( top - cells 71–80 ) , generate the intrinsic gamma oscillations . The onset-response population ( cells 31–45 ) are responsible for resetting the phase of the intrinsic oscillations , thus , synchronizing them to the onset of the external stimulus . Cells in the coding population ( 25 cells , no . 46–70 ) are plotted in an increasing order of their ‘sensitivity’ from bottom ( cell 46 - least sensitive ) to top ( cell 70 - most sensitive ) . The three Figures 2A , 2B , and 2C show the population response to stimuli with three different shape parameter values , and , respectively . For a very fast-rising stimulus ( Figure 2A , ) , cells in the coding population will tend to fire in the first cycle immediately after the onset . For a slower-rising stimulus ( Figure 2B , ) , few cells will fire in the first cycle and most cells will fire in the second cycle after the onset . For a stimulus that rises even slower ( Figure 2C , ) , few cells will fire in the second cycle and most cells will fire in the third cycle after the onset . Thus , intrinsic oscillations discretize the coding population response in the following sense: the external stimulus overlaps approximately three gamma cycles . Every cell can fire at most a single spike during every cycle . The specific spike pattern of every cell depends on its identity ( i . e . , different cells in the coding population have different sensitivity due to different DC input levels ) as well as on the stimulus shape . Hence , the list of which cell fired during what cycle contains information about the stimulus shape . Below we define a binary representation of the neural response that will be used to quantify the information content of the response . We represent the neural response by a binary matrix of size: [number of coding cells]×[three gamma cycles] ( 25×3 in our model ) . Matrix element ( ) indicates whether cell in the coding population fired ( 1 ) or did not fire ( 0 ) in the cycles following the stimulus onset . This choice of binary representation ignores information that may exist on a time scale finer than the gamma cycle . Figure 3 demonstrates the binning procedure ( complete description of the procedure appears in Materials and Methods section , below ) . The mean firing time of the onset population ( plus 4 . 5 ms ) defines the start of the first bin . The boundaries of the bins are defined by the mean spike times of the inhibitory cell population plus 4 . 5 ms ( vertical dotted lines in Figure 3A ) . Figure 3B shows the binary representation of the network response in Figure 3A . The activity of every cell in the coding population during the three gamma cycles in which stimulus is presented is shown by a single row . Every row is divided into three columns that show the firing of the cell during each cycle in black ( fired ) and white ( did not fire ) . The information content can be quantified by measuring the discriminability of the binary representation of stimuli with different shapes . We chose a very simple readout mechanism , based on template matching . Every stimulus is associated with an internal binary template ( see Materials and Methods ) . For a given response , the estimated sawtooth shape parameter is defined as the one associated with the closest template . Hamming distance was used as the distance measure between templates and input response . These choices were made due to their simplicity and the fact that they emphasize the binary nature of the neural responses . Neither the template nor the distance measure was chosen to optimize the estimation accuracy . We do not mean to suggest that the central nervous system uses this particular readout mechanism . Nevertheless , this readout is an appropriate metric for assessing the accuracy of population response in representing sawtooth-shape waveforms . A convenient description of the readout discrimination power is the confusion matrix , ( see Materials and Methods ) . Figure 4 shows the confusion matrix for A three alternative shape parameter values: and B nine alternative shape parameter values: . The probability of a correct classification provides a scalar summary of the of the confusion matrix . In the three alternative tasks , A , the system is always correct , the probability of correct classification is ( chance level is 1/3 ) . In the more difficult nine alternative task B performance decreases , ( chance level 1/9 ) . However , errors in estimating the shape parameter , , have a magnitude: ( where is the estimated shape parameter; see Materials and Methods equation 7 ) . As can be seen from the confusion matrix , although the error rate increases , the errors are small , typically ( the first off-diagonal elements in the confusion matrix ) . Figure 5A shows the the percent correct classification in an alternative ( ) forced choice task , as a function of . For large , the percent correct decays to zero inversely with the number of alternatives , . This results from a finite resolution in the representation of the shape parameter . The confusion matrix in the case of alternatives is shown in Figure 5B . As in Figure 4B , we observe that the confusion matrix has relatively large elements mainly close to the diagonal . Hence , although there is considerable probability of error , the magnitude of the error is typically small . This finite resolution can be quantified by the root mean square ( RMS ) of the estimation error , , where denotes average of over different trials and phase relations . Here we obtain . In order to obtain this resolution a reliable representation is required . Below we show the necessity of the phase resetting mechanism by the onset population for obtaining a reliable representation of the shape parameter . Since network oscillations are intrinsic and the stimulus is external , the oscillation phase at the time of stimulus onset is arbitrary . In the absence of a phase resetting ( synchronizing ) mechanism , the same stimulus may elicit very different responses , depending on exact phase relation . This added variability of the neural responses to the stimulus increases the dispersion of the responses to the same stimulus around the template and can be thought of as added noise . Hence , the templates become less representative and the readout performance decreases . Figure 6 shows the confusion matrix in the three alternative task , , in the absence of the onset signal ( see Figure 4A for comparison ) . As can be seen from the figure , the probability of correct classification decreased dramatically: , relative to , in the case with the onset signal . Nevertheless , performance is still above chance ( chance level is 1/3 ) . It is important to note that the onset signal does not need to precede the stimulus . The requirement is that the onset signal activates the onset population before the coding population responds to the stimulus . In a diphone , typically , onset is shared among all frequency bands; hence , it provides a clear and robust signal . In a recent work Chase and Young [25] have demonstrated how an onset signal can be accurately reconstructed from the response of a population of inferior colliculus cells of the cat and then used to estimate the external stimulus . Thus the onset response assists in stabilizing a reliable representation of the stimulus shape by the neural responses . However , it does not erase all traces of the past . Even with the presence of the onset signal , the neural response to the stimulus depends on the phase relation , but to a smaller extent . This variability in the neural responses to the same stimulus is , in part , responsible for the finite resolution of the representation in the absence of intrinsic noise . Yet another factor that limits the resolution with which the network can represent the stimulus shape is our choice of binary representation . For example , one may imagine two close but different stimuli which elicit neural responses that differ by their exact spike times but fire during the same gamma cycle; these will be indistinguishable in our binary representation . Below we show that this insensitivity to exact spike timing is advantageous in representing time-warped stimuli . Time warp is a very common perturbation in speech signal . A desired property of speech representation is robustness to such perturbations . In order to study the robustness of our representation we modified the stimulus duration and measured our readout performance , keeping the same templates . Figure 7A shows the quality of representation , in terms of percent correct classification in the three alternative task , as a function of the stimulus duration . All network parameters remained unchanged . The templates were obtained from the network response to 50 ms stimulus duration , as in previous sections . As can be seen from the figure , probability of correct discrimination is maximal when the stimulus duration is 50 ms and decreases as the stimulus duration is changed . Nevertheless , there exists a large range of durations 45–75 ms in which probability of correct discrimination is well above chance level . The type of errors caused by time warping of the stimulus depends on the specific time stretch . To see this , it is convenient to further classify errors into three groups: immediate-up , immediate-down and other . In the alternative forced choice task , errors in which stimulus was estimated to be were classified as immediate-up ( down ) . Figure 7B shows the error type distribution as a function of stimulus duration . As in Figure 7A , all network parameters remained unchanged and the templates were obtained from the network response to 50 ms stimulus duration . From the figure , one can see that immediate-down error rate ( blue ) increases when the stimulus duration is increased , whereas immediate-up error rate ( red ) increases when stimulus duration is decreased in the alternative forced choice task . Thus , error type follows the direction of time warping . Figures 7C and 7D show the percent correct and error type distribution as in Figures 7A and 7B , respectively , in the alternative forced choice task . Results in the case are similar to the . Probability of correct discrimination , , peaks at the duration used to obtain the templates , 50 ms , as the stimulus duration is changed , decreases . The immediate-down error rate is increased when stimulus duration is increased and vice versa for immediate-up error rate . Similarly , there exists a range of stimulus durations ( of about 45–65 ms ) for which probability of correct classification is well above chance level . However , this range is smaller for the case than it is for the case . This difference is discussed below . Robustness to time warp comes at the expense of the resolution of the representation . This can be seen by comparing Figures 7A and 7B . When a higher resolution ( alternatives ) is required , the range of durations in which the readout is robust to time warp is decreased , relative to the lower resolution case ( alternatives ) , see above . This notion can be further quantified by studying the RMS estimation error as a function of the amount of time warp of the stimulus . Figure 8 shows the RMS estimation error , , as a function of the amount of time warp of the stimulus duration . As can be seen from the figure , for stimulus durations of 50–70 ms the resolution fluctuates around its maximum ( is minimal ) . The resolution decreases ( increases ) as the amount of time warp increases in its magnitude , both above 70 ms and below 50 ms . All of the above numerical simulations quantifying the network ability to represent time varying stimuli were done in a deterministic model , in the absence of intrinsic noise to the neural dynamics . For example , every inhibitory cell fired during every gamma cycle and every excitatory cell in the gamma generating population fired every other cycle . In a more realistic model [22] , [26] , [27] firing will be sparse and noisy , with oscillations that appear only on the network level . Thus , one should think of every cell in our deterministic model as an “effective cell” , representing the firing of a group of sparsely firing neurons . However , intrinsic noise that may cause spike time jitter , addition or deletion of spikes can have drastic detrimental effect on the quality of a temporal code [28] , [29] . It is therefore important to test the sensitivity of this representation to intrinsic noise . Figure 9 shows the percent correct classification as a function of the input noise level for three , five and nine alternatives ( top to bottom ) . As expected , the probability of correct discrimination is a monotonically decreasing function of noise level . Nevertheless , good performance levels are retained for moderate noise levels . Note , for three alternatives decreased by less than 5% , for five alternatives decreased by 23% and for nine alternatives decreased by 33% . This corresponds to a natural tradeoff of the representation resolution and robustness to intrinsic noise fluctuations .
In our numerical simulations we made certain choices that are required to define the system but are not essential for our qualitative results . We chose to represent the external stimulus by neural responses that extend over internal gamma cycles . The specific choice of gamma cycles is arbitrary and our approach could be easily generalized to cycles . Larger values imply that the stimulus can be represented to a finer resolution . However , finer resolution comes at the expense of robustness to noise and time-warping perturbations . The neurons in our simulations follow Hodgkin-Huxley dynamics ( see Materials and Methods below ) . This choice is also not essential to our main conclusions . Other choices for the neural dynamics , such as integrate and fire , may generate representations that are different in their fine details but still preserve the central qualitative features reported here . Namely: the oscillations discretize the output , forming a binary representation that is robust to moderate levels of noise and time warping perturbations of the external stimulus and is characterized by a tradeoff of sensitivity and robustness . The essential features of our network are the architecture of a PING mechanism for generating the gamma oscillations and the manner in which the external stimulus interacts with the internal oscillations . Speech is an important example of a time-varying signal . There is a natural hierarchy of timescales in speech: phone , diphone , syllable , word , and sentence . The time duration of phones and diphones is on the order of a few gamma cycles , while the duration of a word is roughly that of a theta cycles . Oscillations on different timescales in the auditory cortex have been shown to be organized hierarchically: delta modulates theta , theta modulates gamma [34] . These data support a view of a network with nested oscillations on different timescales [35]–[39] . Though a diphone can be correlated with a beta frequency period or multiple gamma frequency periods , we chose to explore the role of gamma frequency oscillations , since gamma oscillations are known to be prominent in early sensory processing ( see , e . g . , [16]–[20] ) , and to help produce cell assemblies [40] . The nesting of oscillations has a potential relationship to robustness to time warping . Empirical studies of speech [41] , [42] as well as of birdsong [43] have shown positive correlations in time warping fluctuations of short speech and birdsong segments . For example , the degree of time warping of a specific syllable in Zebra finch song can be predicted , to a large extent , by the degree of time warping of previous syllables . Similarly , in speech , time warping fluctuations of nearby short speech segments are correlated . The correlated time stretch can be predicated by estimating a ‘tempo variable’ , such as the prosody , that varies on a longer timescale . Such a tempo variable can be used by an oscillatory network to modulate its oscillation frequency to compensate for the time warp of the stimulus . The mechanism that we suggest for the time encoding lends itself naturally to such a tempo variable , since the PING gamma has increasing frequency with increased drive; any mechanism that can increase drive with faster prosody will produce more robustness to time warp variability of the auditory stimulus . The frequency of a slower but correlated rhythm , such as theta [44] , could act as such a tempo variable . We note that theta rhythms and gamma rhythms sometimes covary in their frequencies [45] . The beta frequency may be associated with the onset signals . In mainstream models of spoken-word recognition the speech waveform is processed by a front-end , providing a representation from which a phonetic transcription is generated . The sequence of phones recognized is then integrated into a form that results in a ‘pointer’ to a specific item in the lexicon . Phonetic transcription is usually accomplished by a search within a vocabulary of acoustic models of the phones . These models are statistical in nature , and the probabilistic model is acquired by training [46] , [47] . While such Hidden Markov Models ( HMMs ) have shown themselves to be highly effective , it is reasonable to question certain properties of their basic structure as a model for biological systems of speech processing . The conditional independence assumption imposed by HMMs is a poor model for the dynamics in the speech signal [48] . It is also extremely difficult to model long-range dependencies with an HMM [49] . Thus , methods which can better model temporal-spectral dynamics inherent in speech are highly desirable . Our long-term goal is to use the physiological aspects of speech processing to improve our understanding of speech representation . In the work discussed here , a first step in this endeavor , we quantify how our model represents a cartoon signal mimicking the response of one cochlear frequency-band to speech input . Many difficult questions have to be answered before we can implement this model as a front-end to a speech recognition system . For example , what is the discrimination power of the model for more realistic signals at the input of a single cochlear channel , e . g . , for a set of signals that are different in shape , in duration , in amplitude ? Can our model provide a stable representation with respect to time scale variations that conform with realistic phonemic variation ( usually not a uniform time warp in nature ) ? How to synchronize an onset signal with the signals across several cochlear channels ( with relative time alignment dictated by the speech source ) ? How to integrate across all cochlear channels ? A system based on the principles of neuronal processing that answers these questions also has the potential to create a paradigm shift in the way that speech is processed by machines . A closely related model was suggested by Hopfield [5] . The focus of this model was on readout of the activity of multiple integrate-and-fire neurons , each of which integrates over time the time-varying signal for a single “channel” . From the perspective of representing speech , the Hopfield model is complete; it suggests an architecture , with a subthreshold gamma oscillator at the core , in which all frequency bands are integrated via a well defined readout mechanism . Although we do not have a complete system yet , a comparison can be made between our model and Hopfield's for a single frequency-band signal . Hopfield used subthreshold oscillations to synchronize the firing across channels , forcing the cells to fire in a “window of opportunity” . Though the equations that embody the model have some memory beyond one cycle , the memory corresponds to a small negative Lyapunov exponent , which is also associated with lack of robustness . Thus , it is unclear how well this performs for a longer time-varying signal . In contrast , our model is not focused on readout , but on representation . The oscillations are used to discretize the signal across several periods , rather than to synchronize many channels . Spike times are determined by both the endogenous gamma rhythm and the external input . This mechanism allows the external stimulus to modulate the frequency of the intrinsic oscillation , unlike the fixed period in the Hopfield model . The idea that a stimulus may be coded by a sequence of firings in discrete epochs has been discussed in the context of olfaction by Bazehnov et al . [50] , [51] . There are two central differences between their work and ours: First , the Bazhenov et al . papers deal with a set of signals that all have the same temporal properties: they have a rise time of 100 ms and a decay time of 200 ms , unlike the sawtooth signals of the current work . Second , in [50] , [51] , the different signals excite different ( possibly overlapping ) sets of cells in the coding population , unlike the signals in the current paper , which all excite the same set of cells , but have different effects on them . Thus , the information in the signals is different from that of the Bazhenov papers and the coding strategy is different , even though both result in discretization . The differences in strategy are appropriate for the differences in the kinds of signals to be encoded: the energy in a given auditory frequency band has a varying temporal structure across the set of signals , for which a sawtooth of different shapes provides a characterization . There is no such structure in olfactory signals . In the current work we did not simulate a neural network implementation of our readout mechanism . How can our readout be implemented ? The approach taken by Hopfield lends itself to a simple readout mechanism based on simultaneity . Since our code has more than one “bit” , a more complex readout mechanism is necessary . There are many suggestions in the literature that might be modified to work for this example [14] , [52] . Stimulus identity , in our model , can be estimated by measuring the time from the firing of the onset cells to the firing of the coding cell . This could be achieved , for example , by an integrator that starts integrating time at the onset response and stops integration at the response of the coding population neurons . Thus , a class of potential readout mechanisms is that of neuronal integrators . Of particular interest is a single cell integrator model of Loewenstein et al . [53] based on slow calcium dynamics in a dendrite of a single cell . In their model [53] , calcium level along the dendrite transitions from high to low and the location of the transition point along the dendrite is determined by integration over time of dendritic inputs . Thus , the firing rate of the cell corresponds to the time integral of the cells' dendritic inputs . Readout of a multiple-bit code might make use of input to multiple dendritic branches . Other ways to estimate such times use long-term potentiation and depression [54] and physiological slow conductances [55] . The above are more appropriate to the current model than the Tempotron [56] , which can distinguish arbitrary time varying inputs , but is unable to discriminate well temporal features that extend beyond its integration time . In this work we studied a very simplified stimulus model . The envelope amplitude of a diphone stimulus in a single frequency channel was approximated by a sawtooth . Incorporating a wider range of envelope repertoire as well as ranges of amplitude and several frequency bands will result in a much richer temporal code and will , most likely , require a larger neural population . However , this richness of detail may impair the clarity of our results . Moreover , meaningful theoretical investigation along these lines requires a better empirical understanding of cortical oscillations during speech perception to yield the essential constraints for theory . For example , when studying a model of several frequency channels we must choose whether or not the onset stimulus and the oscillations are shared among the different channels . Different choices may lead to different results , without reason to choose one over another . The question of whether oscillations are shared is an empirical question . To pursue in a meaningful manner the theoretical framework begun in the current work requires empirical effort to characterize the interaction of neural oscillations with time varying stimuli across several frequency channels . The current framework motivates such empirical work by suggesting ways in which an external stimulus can interact with the dynamics that encodes the signal . | Sensory processing of time-varying stimuli , such as speech , is associated with high-frequency oscillatory cortical activity , the functional significance of which is still unknown . One possibility is that the oscillations are part of a stimulus-encoding mechanism . Here , we investigate a computational model of such a mechanism , a spiking neuronal network whose intrinsic oscillations interact with external input ( waveforms simulating short speech segments in a single acoustic frequency band ) to encode stimuli that extend over a time interval longer than the oscillation's period . The network implements a temporally sparse encoding , whose robustness to time warping and neuronal noise we quantify . To our knowledge , this study is the first to demonstrate that a biophysically plausible model of oscillations occurring in the processing of auditory input may generate a representation of signals that span multiple oscillation cycles . | [
"Abstract",
"Introduction",
"Results",
"Discussion"
] | [
"neuroscience/theoretical",
"neuroscience",
"neuroscience/sensory",
"systems",
"computational",
"biology/computational",
"neuroscience",
"computational",
"biology",
"neuroscience",
"computational",
"biology/systems",
"biology"
] | 2009 | Representation of Time-Varying Stimuli by a Network Exhibiting Oscillations on a Faster Time Scale |
In the early 20th century , there were few therapeutic options for mental illness and asylum numbers were rising . This pessimistic outlook favoured the rise of the eugenics movement . Heredity was assumed to be the principal cause of mental illness . Politicians , scientists and clinicians in North America and Europe called for compulsory sterilisation of the mentally ill . Psychiatric genetic research aimed to prove a Mendelian mode of inheritance as a scientific justification for these measures . Ernst Rüdin’s seminal 1916 epidemiological study on inheritance of dementia praecox featured large , systematically ascertained samples and statistical analyses . Rüdin’s 1922–1925 study on the inheritance of “manic-depressive insanity” was completed in manuscript form , but never published . It failed to prove a pattern of Mendelian inheritance , counter to the tenets of eugenics of which Rüdin was a prominent proponent . It appears he withheld the study from publication , unable to reconcile this contradiction , thus subordinating his carefully derived scientific findings to his ideological preoccupations . Instead , Rüdin continued to promote prevention of assumed hereditary mental illnesses by prohibition of marriage or sterilisation and was influential in the introduction by the National Socialist regime of the 1933 “Law for the Prevention of Hereditarily Diseased Offspring” ( Gesetz zur Verhütung erbkranken Nachwuchses ) .
At the DFA , Rüdin then worked on his second major inheritance study , on the genetic inheritance of “manic-depressive insanity” ( Zur Vererbung des manisch-depressiven Irreseins ) [28] . The original manuscript in the Historical Archives of the Max Planck Institute of Psychiatry in Munich is undated; on the basis of the publications cited the compilation can be dated to the period 1922–1925 [8] . It was never published despite being intended for chapter 4 of “Studies on the Inheritance and Origin of Mental Illness” , one of a series of monographs from the publisher Springer-Verlag of Berlin , covering the full domain of neurology and psychiatry [29] . The Rüdin biographer and psychiatrist Matthias Weber described the study as “the most comprehensive and probably most significant of Rüdin’s works” [8] . The manuscript consists of approximately 160 pages of unbound typescript , each chapter paginated separately , with Rüdin’s hand-written corrections , and numerous large-format , hand-written charts [28] . Some chapters are incomplete and untitled and are not considered in this paper . An additional 250 pages of Rüdin’s hand-written notes , unpaginated , unsystematised and partially in shorthand are mostly illegible . The bibliography is hand-written and only partially intelligible . Rüdin characterised the study as a complement to his 1916 dementia praecox study and used the same methodology [24] . Whilst not excluding a role for environmental factors , he assumed inheritance of a disposition to affective disorders as generally acknowledged but unproven , citing Kraepelin [30] . Rüdin predicted Mendelian dominant inheritance of affective disorders [26] . He intended to prove this on the basis of the segregation pattern in the families of his probands , or determine another mode of Mendelian inheritance . Criteria could then be derived for selection of persons for the eugenic measures he had formulated earlier [20] . The inclusion criteria encompassed all patients admitted as inpatients to the Psychiatric University Clinic of Munich ( Psychiatrische Universitätsklinik München ) with a diagnosis of “manic-depressive insanity” since 1904 . No preference was given to patients with a dense family history; the diagnosis alone was the criterion for inclusion . This type of systematic ascertainment was unusual in psychiatry at a time when spectacular single case studies or densely affected families were the main focus of research [31 , 32] . Follow-ups were conducted to evaluate the course of the illness and verify the diagnosis . Rüdin sought accurate clinical diagnostics for inclusion and exclusion of probands , using Kraepelinian nosology , which was widely recognised though not without controversy [33 , 34] . Under the diagnostic rubric of “manic-depressive insanity” , Kraepelin included “periodic and circular insanity” , “simple mania” , “melancholy” and “amentia” [30] . This corresponds approximately to the modern-day categorisation of affective disorders [35–38] . For the diagnosis Rüdin depended above all on the clinical symptoms and less on the course of the disease . Kraepelin regarded “manic-depressive insanity” as “curable” and dementia praecox as “incurable” . Rüdin emphasised however that cases of both disorders were documented that did not follow this rule . He called upon a second clinician with no knowledge of the family history to verify the diagnoses [28] . The sample were hospitalised patients , thus biasing to more severe cases [15] . The patients and , whenever possible , family members were interviewed; further sources included medical and administrative files . However , when a medical diagnostic interview could not be arranged for family members , these diagnoses were often based on descriptions of maladjusted or otherwise conspicuous family members , how often this occurred is not recorded . The most important information ( age , diagnosis , medical history and family history ) was noted on standard cards , as implemented by Kraepelin , and in detailed family trees [24] . Rüdin integrated Weinberg’s statistical methods and Mendelian genetics into a research methodology which sought to predict the passing on of mental illnesses within families: the “empirical heredity prognosis” . His second goal was to prove a Mendelian mode of inheritance via the systematic ascertainment of as many patients with affective disorders as possible . Rüdin recognised the problem that in recessive modes of inheritance , only families in which the disposition was expressed could be counted , while families with heterozygote carriers were lost . In order to be able to apply Mendelian rules to a sample , the ascertained morbidities would also ideally have to correspond to those in the population . Rüdin applied Weinberg’s simple sib and proband method to that end [25] . He stayed in close contact with Weinberg during the evaluation of his results; their correspondence is cited in several places in the manuscript . The overrepresentation of patients with affective disorders in the sample was corrected by the simple sib method , whereby index cases , referred to as “probands” ( Probanden ) , were excluded from the calculation . Affected siblings were entered into the calculation as “secondary cases” ( Sekundärfälle ) . The proband method was used to correct for multiple ascertainment in families with several probands . Rüdin also applied two different methods for age correction , allowing for the possibility that a family member who was healthy at interview could become ill later [28 , 39] . On the basis of age distribution , Rüdin determined the period of risk for the onset of affective disorders to be between the ages of 14 and 68 . The number of “lifetimes at risk” ( Bezugsziffer ) was then calculated from the sum of siblings , corrected for their age ( Table 2 ) . Rüdin’s “empirical heredity prognosis” was based on the calculation of a predictive risk of illness , the “morbid risk” ( Morbiditätsrisiko ) , which included the above-mentioned Weinberg methods . The morbid risk was calculated from the number of affected persons in the sample , corrected by the simple sib and proband methods , divided by the number of lifetimes at risk , that were calculated by the age correction method ( Table 2 ) . Calculating the morbid risk , Rüdin assumed full penetrance for the inherited traits in question and excluded a possible influence of external factors or interference with other genes . The thus calculated morbid risk was compared with the proportions expected from a Mendelian crossing in order to prove Mendelian inheritance and thereby the inheritance of affective disorders as such . Rüdin also used morbid risk as a predictive value for any particular person with certain preconditions ( e . g . a parent with an affective disorder ) to develop an affective disorder at some point in their life , and therefore serve as a kind of “genetic counseling” , or as he called it , an “empirical heredity prognosis” [28] . Rüdin’s use of statistical processes was a groundbreaking approach at the beginning of the 20th century and led to sound results that in part remain valid today [32 , 40] . However , while focussing on the methods of Weinberg , who was also chairman of the “Society for Racial Hygiene” and , like Rüdin , a staff member of the “Archive of Racial and Social Biology” , Rüdin ignored some other statistical methods , such as correlational analyses , which were in common use [15] . After excluding unconfirmed diagnoses , the sample comprised 661 probands from 650 families , or “sibships” , comprised of 4351 siblings in total ( Fig 1 ) . In 566 families both parents were healthy and in 84 families one parent had an affective disorder ( Table 3 ) . Rüdin analysed these two groups independently ( Table 4 ) . Consideration was given to other aspects such as alcoholism in the parents or mental illness in other relatives , in which case the families’ standard cards were re-sorted according to the question posed , and the proportions recalculated . The morbid risk was then calculated for various categories , and Rüdin found an increase of affective disorders in children of parents who had affective disorders . Contrary to the widely held assumption of a high inheritance rate of affective disorders [30 ) , Rüdin calculated a substantially lower proportion with 7 . 43% affected children of healthy parents and 23 . 82% affected children of an affected parent . For an overview of Rudin’s study results see Table 5 . Weinberg calculated the proportions expected from crossing different genetic strains , which were then compared with the morbid risks determined by Rüdin in order to find as close a match as possible . Because a simple recessive or dominant inheritance had to be excluded , Rüdin , with the aid of Weinberg’s calculations , took more and more complicated modes of Mendelian inheritance into consideration . Instead of looking more closely at other factors such as patients’ living conditions , comorbidity or other external influences , calculations with assumptions of up to 12 interacting alleles were made in order to come as close as possible to the expected proportions . With the assumption of a three-locus model with two recessive and one dominant factor , Rüdin’s proportions ultimately best matched those calculated by Weinberg , which is why he postulated this mode of inheritance as the most probable ( Table 6 ) . However , implementation of the “empirical heredity prognosis” was unsuccessful; it was still not possible to use a standardised formula to predict the probability that a particular individual would develop an illness . Hence the way one should conceive the genotype of a certain patient was still far from clear . Rüdin emphasised the preliminarity of his results and called for further studies with larger sample size and avoidance of assortative mating . In order to better measure the effects of environmental factors , Rüdin initiated further family and twin studies into the inheritance of affective disorders at the GDA [41–43] . Aubrey Lewis ( 1900–1975 ) , in 1934 , considered Rüdin’s studies the starting point in the field , refuting allegations against psychiatric genetics “that it was bad psychiatry and bad genetics” [44] . Lewis distinguished between Rüdin’s research and demands he made for changes in health policy , sharply criticising the eugenic measures in Germany’s “Law for the Prevention of Hereditarily Diseased Offspring” , albeit anonymously [45 , 46] . In German-speaking psychiatry , there were well-known critics of Rüdin’s eugenic demands on society; important psychiatrists like Karl Jaspers ( 1883–1969 ) , Oswald Bumke ( 1877–1950 ) and Eugen Bleuler ( 1857–1939 ) argued the mode of inheritance was unproven; external and social factors influenced disease course , and the right to personal self-determination had to be respected [15 , 47–51] ( for further insight into resistance of medical professionals against negative eugenic measures , see [52–54] ) . Rüdin’s concept of an empirical heredity prognosis served as a methodological model for many subsequent studies at the DFA , known as the Munich School [55–60] . European and American scientists , some of whom had been fellows of the Genealogic-Demographic Department ( GDA ) of the DFA , used Rüdin’s research methodology in psychiatric genetic studies [27 , 27 , 60–65 , 65–69] . The results of the twin and family studies undertaken at the Munich School remained valid in their methodology and results for decades [31 , 42 , 70–72] . Whilst Rüdin actively published over decades [8] , only one article reported the methodology he prized , namely his large-scale study on the inheritance of dementia praecox [24] , with which he acquired his reputation as a psychiatric geneticist [15] . It is all the more surprising that the similarly laid out study on “manic-depressive insanity” , which was elaborately prepared and carried out over the course of years was never published [8 , 28] . Contributing factors may include Rüdin , concentrating on his political career , paid less attention to the practical implementation of his research plans [15] . This is somewhat contradicted , by the fact that the study and manuscript were essentially completed . There is some evidence that Rüdin doubted his results . He concurred with eugenicist and heredity theorist Ludwig Plate ( 1862–1937 ) , that diagnostic uncertainty necessitated skeptical application of Mendelian rules [73–75] . Most likely is that his demands for negative eugenic measures against patients with affective disorders and their families could not be justified on the grounds of the heredity figures he had calculated . With the benefit of hindsight , the inheritance figures Rüdin calculated have been confirmed repeatedly [71 , 76] , and the search for replicable gene variants leading to the onset of affective disorders continues [77] . In a 1924 lecture Rüdin even recognised that environmental factors combined with disposition to illness , trigger onset of disease [78] . Rüdin’s actual results were therefore reasonably sound from today’s perspective , notwithstanding methodological limitations and that his studies were undertaken prior to knowledge of DNA , the double helix , and the intricacies of molecular genetics , epigenetics and endophenotypes [79–84] . Selective publication of positive results remains contentious today . The German human geneticist Peter Propping considers it the greatest danger to psychiatric genetics; “a silent coalition exists between an author and an editor: both are interested in publishing positive findings” [40] . He calls for a platform where all relevant results are accessible to the scientific public , to minimise bias in publishing [40] . In contrast to Propping , prominent scientists like Christiane Nüsslein-Volhard who received the Nobel Prize in Physiology or Medicine in 1995 recommend not publishing negative results , because science should increase knowledge , not merely produce more data . In a recent interview she points to methodological errors as a potential reason for negative results [85] . Rüdin too may have been dubious about his results and therefore refrained from publishing the study . Rüdin had already clearly formulated his research aim before his study began . He unethically promulgated his eugenic ideology based on a selective and at times patently false reading of his results or even ignoring them [40 , 86] . For further discussion on publication bias , see [87 , 88] . In 1933 , Rüdin chaired the committee for racial hygiene and racial policy at the ministry of the interior of the NS regime and collaborated on the “Law for the Prevention of Hereditarily Diseased Offspring” [8 , 17] . For more detailed information about the role that scientists played in bolstering the racial theories of the NS regime see [3] . According to this law , all persons who had been determined to be hereditarily diseased according to medical science were to be compulsorily sterilised [17] ( for the origins of the law see [8] ) . In his commentary on the law’s implementation , Rüdin justified sterilising psychiatric patients based on the results of his study of the inheritance of dementia praecox , extrapolating the allegedly proven inheritance to other psychiatric and neurologic diagnoses [15 , 17 , 89 , 90] . In doing so , Rüdin reinterpreted his results as much more conclusive and reliable than he had in earlier commentaries , but without citing any later studies [24 , 74] . As the story of the 1933 sterilisation-law shows , Rüdin’s research results were accepted uncritically as a scientific basis for legislation , strengthening the case for the National Socialists’ health and social policy . Between the law coming into force in 1934 , until 1945 , between 350 , 000 and 400 , 000 persons were sterilised [91–93] . The DFA , founded by Kraepelin and later led by Rüdin , was among the Institutes that issued registration forms required for sterilisation of patients [13] . The number of sterilisations only decreased when , in 1940 and 1941 , the Nazis progressed to killing mentally ill and handicapped patients , in “Aktion T4” , under the euphemism “euthanasia” [94] ( for the history of eugenics and euthanasia , see also [95–97] ) . Rüdin became aware of this secret operation at the end of 1939 at the latest , but was not directly involved in its preparation or execution [8] . However , his backing may have influenced important decisions at the highest political levels in favour of killing patients [98 , 99] . Rüdin supported research projects which included killing designated patients for post mortem material [100 , 101] . Also the meticulously recorded registration of patients in Rüdin’s studies later facilitated locating the victims for forced sterilisation and “Aktion T4” [98 , 102] . Because his scientific interests were so consistent with Nazi ideology , the DFA was supported by the various centers of power in the National Socialist state [96 , 103] . Rüdin did not consider his associations with these ideologies to compromise the scientific quality of his empirical studies [16] . In celebration of the tenth anniversary of the National Socialist state , Rüdin composed a laudation in praise of Adolf Hitler’s services to racial hygiene [104] . In a memorandum he referred to “euthanasia” as a component of “therapeutic reform” [10 , 105] . Contemporary views of Rüdin’s work diverge widely , and are subject to ongoing controversy [13 , 90 , 99 , 106] . Some would strip him entirely of his rank as a scientist [15]; others credit him with the establishment of modern psychiatric human genetics through the development of the “empirical heredity prognosis” [8 , 31 , 40] . Appeals to discontinue citing Rüdin’s scientific work have been put forward [107] . Certainly any scientific analysis of Rüdin’s works should provide information about his political role [108] . His legacy is deeply troubling but highly illustrative of the nexus between science , ideologies , ethics and humanity . | Psychiatric genetics was established as a scientific discipline in the early 20th century . The intention was to provide evidence for the inheritance of mental illnesses . This was to open up new paths in prevention , since therapeutic options at the time were meagre . Applying modern study designs and statistical methods , the German psychiatrist Ernst Rüdin ( 1874–1952 ) found lower inheritance of affective disorders than anticipated . He surmised that external factors were also important in their development . From the vantage point of the present , his results were sound . However , this major study , titled “On the Inheritance of Manic-Depressive Insanity” , though developed as a manuscript , was never published . The question arises whether Rüdin doubted his own scientific methods and hence results , or if he was influenced by his ideological views and the expectations of the scientific community and German society . Whilst withholding his results , he continued to promote prevention of assumed hereditary mental illnesses by prohibition of marriage or sterilisation . | [
"Abstract",
"Introduction"
] | [] | 2015 | Ernst Rüdin’s Unpublished 1922-1925 Study “Inheritance of Manic-Depressive Insanity”: Genetic Research Findings Subordinated to Eugenic Ideology |
Single-cell mRNA sequencing can uncover novel cell-to-cell heterogeneity in gene expression levels in seemingly homogeneous populations of cells . However , these experiments are prone to high levels of unexplained technical noise , creating new challenges for identifying genes that show genuine heterogeneous expression within the population of cells under study . BASiCS ( Bayesian Analysis of Single-Cell Sequencing data ) is an integrated Bayesian hierarchical model where: ( i ) cell-specific normalisation constants are estimated as part of the model parameters , ( ii ) technical variability is quantified based on spike-in genes that are artificially introduced to each analysed cell’s lysate and ( iii ) the total variability of the expression counts is decomposed into technical and biological components . BASiCS also provides an intuitive detection criterion for highly ( or lowly ) variable genes within the population of cells under study . This is formalised by means of tail posterior probabilities associated to high ( or low ) biological cell-to-cell variance contributions , quantities that can be easily interpreted by users . We demonstrate our method using gene expression measurements from mouse Embryonic Stem Cells . Cross-validation and meaningful enrichment of gene ontology categories within genes classified as highly ( or lowly ) variable supports the efficacy of our approach .
Current technology allows the analysis of gene expression with high resolution . Instead of measuring average expression levels across a bulk population , scientists can now report information at the single cell level using techniques such as single-cell RNA-sequencing ( scRNA-seq ) [1] . Unlike bulk experiments , scRNA-seq can uncover heterogenous gene expression patterns in seemingly homogeneous populations of cells [2] , opening the door to important biological questions that remain otherwise unanswered . However , besides experimental challenges such as the isolation of single cells and parallel sequencing of multiple cDNA libraries [3] , statistical analysis of single-cell level data is itself a challenge [4] . Firstly , cell-specific measurements can vary in scale due to differences in total cellular mRNA content [5] . For instance , in Fig 1 ( a ) , each gene has the same expression rate in both cells , yet the expression counts in the first cell will be roughly twice as much as those from the second cell . In the same spirit , if different sequencing depths ( the number of times a single nucleotide is read during the sequencing ) are applied to these cells , the scale of expression counts will also be affected . Thus , normalisation is a crucial issue in this context . Another fundamental problem for interpreting single-cell sequencing is the presence of high levels of unexplained technical noise ( unrelated to sequencing depth and other amplification biases ) [5] . This creates new challenges for identifying genes that show genuine biological cell-to-cell heterogeneity—beyond that induced by technical variation—and motivates the systematic inclusion of spike-in genes in single-cell experiments . Quantifying genuine heterogeneity in gene expression is an important step as it can lead to the discovery of co-expressed genes and novel cell subpopulations , among others [4][6] . Recently , the introduction of Unique Molecular Identifiers ( UMI ) attached to each cDNA molecule during reverse transcription has substantially reduced the levels of unexplained technical noise and eliminated the effect of sequencing depth changes and other amplification biases in single-cell experiments . Unlike most scRNA-seq datasets published to date—where expression counts likely correspond to the number of reads mapped to each gene—UMI based datasets are recorded in terms of the number of molecules , producing a meaningful scale for the expression counts . Nevertheless , our analysis of a mouse Embryonic Stem Cells ( ESC ) suggests that unexplained technical variability can not be completely removed by using UMIs ( see Results section ) and that an accurate quantification of technical variability still remains important . Throughout , we motivate our method using UMI-based expression counts . However , the methodology described here is general and can be also extended to traditional scRNA-seq experiments ( where expression counts represent the number of short reads mapped to specific genes ) by modifying the interpretation of some model parameters . Typical UMI based scRNA-seq data can be represented by a q × n matrix whose entries are the number of mRNA molecules mapped to specific genes ( proxy for gene expression ) for each cell . More specifically , let Xij be a random variable representing the expression count of a gene i in cell j ( i = 1 , … , q; j = 1 , … , n ) . Thus , in a homogeneous population of cells where the true concentration of fragments from a gene i is μi ( in a suitable unit ) and where measurements are not affected by unexplained technical error , Xij follows a Poisson distribution with rate ϕj sj μi , where ϕj adjusts the expression rate in terms of total mRNA content in cell j and sj accounts for changes in capture efficiency across cells ( for read-based expression counts , the latter also captures differences in sequencing depth and other amplification biases ) . Nonetheless , the Poisson model often predicts smaller variability than is observed in real datasets [7] . This so-called overdispersion is potentially linked to genes whose expression has a substantially larger biological cell-to-cell variability than would be expected in a homogeneous population of cells . However , this excess of variability may also arise from unexplained technical noise [6] . Non-biological spike-in genes ( which are added to the lysis buffer and thence present at the same level in every cell ) can be used to quantify technical noise ( differences in capture efficiency and other unexplained sources ) . A typical example is the set of 92 extrinsic molecules derived by the External RNA Controls Consortium ( ERCC ) [8] . As the number of spike-in molecules added to each cell is known from experimental information , this provides a gold standard to which empirical measurements of spike-in genes’ expression can be compared , enabling a quantitative calibration of the technical noise . Similar strategies have also been used in the context of measurement error problems , where a validation error free group or gold standard measurements provide information about unknown sources of error ( e . g . [9] ) . Consistent with previous related literature ( e . g . [5] , [7] ) , we introduce a model based on a Poisson structure . In BASiCS ( Bayesian Analysis of Single-Cell Sequencing data ) , a joint model of biological and spike-in genes is formulated to simultaneously quantify unexplained technical noise and cell-to-cell biological heterogeneity using the complete set of data , borrowing information between both sets of genes ( spike-in and biological ) through common parameters in a hierarchical structure . Additionally , BASiCS incorporates an automated normalisation method , where normalising constants are treated as model parameters . These constitute major methodological advantages over previous 3-step approaches , where first datasets are pre-normalised and secondly unexplained technical noise is estimated using only the spike-in genes , before these parameters are plugged in when modelling biological data ( ignoring the uncertainty related to the technical fit ) .
Throughout , we analyse the expression counts of q genes , where q0 are expressed in the population of cells under study ( biological genes ) and the remaining q−q0 are spike-in ( technical ) genes . Let Xij be a random variable representing the expression count of a gene i in cell j ( i = 1 , … , q; j = 1 , … , n ) . Firstly , we define a model for the technical genes , whose expression counts are not affected by total cellular mRNA content ( see Fig 1 ( b ) ) , thus the cell-specific size factors ϕj are not required ( in this case , the normalisation must only account for differences in capture efficiency via the sj’s ) . Naturally , for spike-in genes , deviations from a Poisson formulation are due only to unexplained technical variability . We assume that this unexplained technical noise depends on cell-specific characteristics and that , for a given cell , it affects the expression counts of all genes in the same manner . Under this assumption , unexplained technical noise can be incorporated through the following hierarchical model X i j | μ i , ν j ∼ ind Poisson ( ν j μ i ) , ν j | s j , θ ∼ ind Gamma ( 1 / θ , 1 / ( s j θ ) ) , i = q 0 + 1 , … , q ; j = 1 , … , n , ( 1 ) where μi represents the normalised expression rate of gene i in the population of cells under study and the random effect νj ( with E ( νj∣sj , θ ) = sj and Var ( ν j ∣ s j , θ ) = s j 2 θ ) fluctuates around the capture efficiency normalising constant sj , quantifying unexplained technical noise via a single hyper-parameter θ , borrowing information across all genes and cells ( see Fig 2 ) . The model in Eq ( 1 ) is equivalent to a negative binomial distribution for the expression counts ( like in [7] ) . In order to accommodate the biological genes , BASICS extends the model in Eq ( 1 ) as X i j | μ i , ϕ j , ν j , ρ i j ∼ ind { Poisson ( ϕ j ν j μ i ρ i j ) , i = 1 , … , q 0 , j = 1 , … , n ; Poisson ( ν j μ i ) , i = q 0 + 1 , … , q , j = 1 , … , n ( 2 ) with ν j | s j , θ ∼ ind Gamma ( 1 / θ , 1 / ( s j θ ) ) and ρ i j | δ i ∼ ind Gamma ( 1 / δ i , 1 / δ i ) , ( 3 ) where νj’s and ρij’s are mutually independent random effects and the cell-specific size factors ϕj are introduced to normalise the biological expression counts according to differences in total mRNA content ( see Fig 1 ( a ) ) . As in Eq ( 1 ) , the νj’s capture cell-to-cell unexplained technical variability , oscillating around the capture efficiency normalising constants ( sj ) according to the strength of unexplained technical variability ( θ ) . The additional random effects , ρij ( with E ( ρij∣δi ) = 1 and Var ( ρij∣δi ) = δi ) , relate to heterogeneous expression of a gene across cells , quantifying biological cell-to-cell variability via gene-specific hyper-parameters δi , borrowing information across all cells ( see Fig 2 ) . Unlike previous stepwise approaches ( e . g . [5] ) , BASiCS treats cell-specific normalising constants ( ϕj’s and sj’s ) as model parameters , and estimates them by combining information across all genes ( see Fig 2 ) , providing simultaneous inference with all other model parameters . Here , the marginal distribution of the expression count of gene i in cell j ( integrating out the random effects νj’s and ρij’s ) induces the same expected counts as in [5] . In fact , E ( X i j | μ i , δ i , ϕ j , s j , θ ) = ϕ j I i s j μ i , with I i = 1 when i ≤ q 0 and I i = 0 otherwise . ( 4 ) In addition , the variance of these expression counts can be decomposed as Var ( X i j | μ i , δ i , ϕ j , s j , θ ) = ϕ j I i s j μ i + θ [ ϕ j I i s j μ i ] 2 + I i δ i ( θ + 1 ) [ ϕ j I i s j μ i ] 2 . ( 5 ) The first term in Eq ( 5 ) is the biological baseline variance—based on a Poisson ( ϕ j I i s j μ i ) model . The second component represents the variance inflation due to unexplained technical variability and the final term is linked to biological cell-to-cell heterogeneity . The decomposition in Eq ( 5 ) is similar ( as a function of the expected counts ) to those proposed in [5] and [7] , which have been validated empirically . Intuitively , highly variable genes ( HVG ) are those for which a large fraction of the total expression variability is explained by a biological cell-to-cell heterogeneity component . Here , we characterise highly variable genes as those for which σ i ≡ δ i ( θ + 1 ) [ ( ϕ s ) * μ i ] - 1 + θ + δ i ( θ + 1 ) > γ H , where ( ϕ s ) * = median j ∈ { 1 , … , n } { ϕ j s j } , ( 6 ) i . e . when the proportion of the total variability of the expression counts of gene i in a reference cell ( derived from Eq ( 5 ) , replacing ϕj sj by ( ϕs ) * in order to represent a typical cell within the analysed sample ) that is explained by biological cell-to-cell heterogeneity exceeds a variance contribution threshold γH . In other words , we characterise as HVG those whose biological cell-to-cell heterogeneity component explains γH × 100% of the total variability ( in a typical cell ) . The latter criterion induces contours in terms of δi , which are given by δ i > [ γ H 1 - γ H ] [ ( ( ϕ s ) * μ i ) - 1 + θ 1 + θ ] . ( 7 ) Naturally , the contour in Eq ( 7 ) is an increasing function of γH . Additionally , it is a decreasing function of the normalised expression rate μi , which is a welcome feature ( previous studies have shown evidence of lower levels of biological cell-to-cell heterogeneity in highly expressed genes [5] ) . BASiCS quantifies the evidence in favour of a gene being highly variable in terms of the upper tail posterior probabilities ( associated to high biological cell-to-cell heterogeneity components ) and labels as HVG those genes such that ( for a given evidence threshold αH ) π i H ( γ H ) = P ( σ i > γ H | { x i j : i = 1 , … , q , j = 1 , … , n } ) > α H , ( 8 ) i . e . when such evidence is strong . Analogously , lowly variable genes ( LVG ) would be those for which π i L ( γ L ) = P ( σ i < γ L | { x i j : i = 1 , … , q , j = 1 , … , n } ) > α L , ( 9 ) for a given variance contribution threshold γL and an evidence threshold αL . Estimates of these quantities can be easily computed based on a posterior sample of the model parameters , requiring minimal computational effort ( other criteria , such as Bayes Factors , usually require intensive calculations [10] ) . Tail posterior probabilities have also been used in the context of differential expression for microarray experiments [11] , providing richer and more interpretable output than standard hypothesis testing procedures . Our method for detecting highly ( and lowly ) variable genes requires the choice of variance contribution thresholds γH and γL as well as evidence thresholds αH and αL . If there is biological motivation behind particular values of γH or γL , these values can be fixed prior to the analysis . However , αH and αL have a technical role , quantifying the uncertainty associated with the detection of HVG and LVG . For fixed values of γH and γL , we can choose optimal values for αH and αL as those where the expected false discovery rate ( EFDR ) and expected false negative rate ( EFNR ) coincide . For the rule in Eq ( 8 ) , these quantities are defined as in [12] and respectively given by EFDR α H = ∑ i = 1 q 0 ( 1 - π i H ( γ H ) ) I ( π i H ( γ H ) > α H ) ∑ i = 1 q 0 I ( π i H ( γ H ) > α H ) and EFNR α H = ∑ i = 1 q 0 π i H ( γ H ) I ( π i H ( γ H ) ≤ α H ) ∑ i = 1 q 0 I ( π i H ( γ H ) ≤ α H ) . ( 10 ) where I ( A ) = 1 if A is true , 0 otherwise . Equivalent expressions can be determined for Eq ( 9 ) , replacing π i H ( γ H ) and αH by π i L ( γ L ) and αL , respectively . Alternatively , if there is no clear pre-determined choice for γH and γL , choosing a specific common value for the EFDR and the EFNR ( e . g . EFDR = EFNR = 10% ) can define optimal values for αH and αL as well as for γH and γL . Beyond the choice of particular thresholds for the detection of highly and lowly variable genes , a key advantage of our method is the generation of a natural ranking of the genes in terms of the percentage of variance explained by the biological cell-to-cell heterogeneity component ( σi ) . For particular threshold choices , our method classifies as highly variable those genes for which σi is high ( above the variance contribution threshold γH ) and where there is strong evidence to support this fact ( the probability of {σi > γH} is above the evidence threshold αH ) . As a result , BASiCS aims to identify key drivers of cell-to-cell heterogeneity rather than complete enumeration . Our analysis does not imply that all genes located below these thresholds have stable expression among the analysed cells . Without additional assumptions , the parameters of the model presented in Eq ( 2 ) and Eq ( 3 ) cannot be identified . However , the cell-specific capture efficiency normalising terms sj’s can be identified if we assume that μq0+1 , … , μq are known . This is not a limitation , because the true concentration of the spike-in genes added to each cell are known from experimental information . In addition , δi’s ( quantifying gene-specific biological cell-to-cell heterogeneity ) and θ ( quantifying unexplained technical variability ) can be identified via the variance of the biological and technical expression counts . Nonetheless , the scale of the ϕj’s ( cell-specific mRNA content normalisation ) is arbitrary because μ1 , … , μq0 are unknown . A simple solution is to impose the restriction n − 1 ∑ j = 1 n ϕ j = ϕ 0 , which can be achieved by reparameterising the model in terms of κ1 , … , κn with ϕ j = ϕ 0 e κ j ∑ j = 1 e κ j , j = 1 , … , n κ 1 = 0 . ( 11 ) Although this restriction imposes an arbitrary scale to the ϕj’s , this does not affect inference about relative differences between the μi’s , nor the δi’s . Therefore , standard analyses , such as the detection of highly variable genes or differential expression , are not affected by particular values of ϕ0 . For simplicity , we recommend ϕ0 = n ( this value will be used hereafter in this article ) . We assume prior independence between all model parameters , using a flat non-informative prior for the normalised expression rates μ1 , … , μq0 and proper informative prior distributions for all other model parameters . Under this prior , Bayesian inference is implemented using an Adaptive Metropolis ( AM ) within Gibbs Sampling ( GS ) algorithm [13] . This algorithm was implemented using a combination of C++ and R via the Rcpp library [14] . An R package has been prepared and is available at: https://github . com/catavallejos/BASiCS More details about the prior specification and the implementation of posterior inference can be found in S1 Text and S2 Text , respectively . Information regarding the computational cost of our method is displayed in S7 Text . Here , we briefly discuss the 3-step method described in [5] to analyse scRNA-seq data and to detect HVG in the population of cells under study ( notably , BASiCS not only provides a method for HVG detection , but LVG can also be identified ) . This method pre-normalises the expression counts using the method available in DESeq [15] , calculating two separate sets of normalising constants as: ω j B = median i = 1 , … , q 0 { x i j ( ∏ j = 1 n x i j ) 1 / n } and ω j T = median i = q 0 + 1 , … , q { x i j ( ∏ j = 1 n x i j ) 1 / n } , j = 1 , … , n ( 12 ) for biological and technical genes , respectively ( in Eq ( 12 ) , xij represents the observed counts of a gene i in cell j ) . In terms of our notation , ω j B and ω j T play the role of ϕj sj and sj , respectively . Based on point estimates of these quantities , normalised expression counts are then computed as x i j * = x i j / ω j B and x i j * = x i j / ω j T for biological and technical genes , respectively . When a large number of genes is being analysed , the variance associated to estimators in Eq ( 12 ) is negligible . However , the expressions in Eq ( 12 ) are undefined if one or more of the expression counts of any analysed gene are equal to zero ( the geometric means in the denominators are equal to zero ) . A common solution is to exclude those genes with zero counts from the normalisation calculations ( but not from any other downstream analysis ) . As a result , these estimators become highly unstable , especially for strong levels of technical noise ( where a high proportion of zero counts is typically observed ) . This is illustrated in Fig 3 ( see panels ( a ) and ( b ) ) , where we simulated data using the same structure as the mouse ESC dataset analysed in the Results section , using the model implemented in BASiCS and a range of values for θ ( including θ = 0 , where there is no unexplained technical noise ) . Fig 3 ( b ) also shows that the stepwise approach proposed in [5] does not recover the correct scale for the sj’s ( not surprising as their method was not designed to do so ) . In contrast , Fig 3 ( c ) shows the superior performance of our approach . This is not surprising because: ( i ) our estimates used the actual expression rates of the spike-in genes ( given by the number of mRNA ERCC molecules added to the lysis buffer of each cell ) , instead of their empirical counterparts ( recovering a correct and meaningful scale for the sj’s ) and ( ii ) we combined information from all genes ( biological and technical ) without having to exclude genes where one or more cell-specific counts were equal to zero . Using the pre-normalised expression counts , [5] proposes a HVG detection method based on the relationship between gene-specific sample means and the corresponding coefficients of variation . An initial fit of this relationship is made using only the spike-in genes ( where heterogeneous expression is only due to a technical component ) , quantifying the effect of unexplained technical variability . The output of this technical fit is then plugged in when modelling biological data , characterising as HVG those whose expression variability substantially exceeds what would be expected due to technical variability ( i . e . the level predicted by the technical fit ) —ignoring the uncertainty associated to the technical fit .
To illustrate BASiCS we consider scRNA-seq data for 7 , 941 genes ( 7 , 895 biological and 46 ERCC spikes ) from 41 mouse ESCs . This corresponds to a subset of the dataset presented in [16] , generated by discarding those genes with total count ( across all cells ) below 41 ( i . e . where the counts are , on average , less than 1 molecule per cell ) . By doing this , we exclude genes with very low expression rates , which have less biological relevance . As illustrated in [16] , the use of UMIs ( attached to each cDNA molecule during reverse transcription ) reduces the strength of technical noise . Nevertheless , our analysis suggests that unexplained technical variability has not been completely removed by this technology ( see discussion below ) . S3 Text describes the input parameters used for the implementation ( including prior hyper-parameters values ) . The data and code used for the analysis are provided in S1 Data . Fig 4 summarises posterior inference for the cell-specific normalising terms ϕj’s and sj’s . Panel ( a ) suggests there is a substantial heterogeneity in the total mRNA content per cell ( ϕj ) and a relatively good correspondence between our estimates and the ones produced by the method in [5] . In the context of UMI datasets , the sj’s can be understood as a measure of changes in capture efficiency . In the ideal case , all the sj coefficients should approach 1 ( i . e . all gene molecules are captured ) . Instead , in the case of the analysed mouse ESCs , the posterior medians of the sj’s vary between 0 . 31 and 0 . 44 across cells ( see panel ( b ) ) , suggesting that part of the original molecules are lost throughout the experiment ( this is particularly critical for lowly expressed genes , as they might not be captured at all ) . As shown in panel ( b ) , the BASiCS estimations of the sj’s show good concordance with the empirical proportions of total spike-in molecules captured in each cell . The small scale difference between the posterior medians of the sj’s and these empirical proportions is due to a highly skewed posterior distribution of the sj’s; however posterior modes closely match these values ( see panels ( c ) and ( d ) ) . Panel ( b ) also shows a strong discrepancy between the methods when estimating the sj’s . Our method suggests that the scale of the technical counts does not substantially vary among cells , which is more reasonable when analysing UMI-based counts . Finally , an important feature of our method is a direct quantification of the uncertainty related to estimation of all normalising constants ϕj’s and sj’s ( by means of high posterior density intervals ) , an element that was ignored in [5] . Despite the use of UMIs , posterior inference strongly suggests the presence of unexplained technical noise in gene expression measurements ( see Fig 5 ) . In fact , the posterior distribution of the unexplained technical variability parameter θ is concentrated away from zero ( see panel ( b ) ) . In addition , even though the posterior distribution of the cell-specific normalising terms sj is homogeneous across cells , panel ( a ) shows substantial differences among the cell-specific random effects ( νj ) . Overall—across all genes—the unexplained technical component explains approximately 28% of the total variability of expression counts in a typical cell . The data also exhibit strong evidence of biological cell-to-cell heterogeneity . In fact , in the case of the analysed mouse ESC dataset , the posterior median of σi ( defined in Eq ( 6 ) ) is above 62% for 50% of the 7 , 895 biological genes ( see Fig 6 ( a ) ) . In addition , Fig 6 ( b ) shows a strong relationship between the biological cell-to-cell heterogeneity ( δi ) and the gene-specific expression rates ( μi ) which is coherent with the contours in Eq ( 7 ) that are decreasing functions of μi . In practice , we define variance contribution thresholds ( γH and γL ) and evidence thresholds ( αH and αL ) for the detection of HVG and LVG by setting the EFDR and the EFNR ( defined as in Eq ( 10 ) ) equal to 10% ( see Table S1 in S4 Text ) . Using this rule , we obtain γH = 0 . 79 , γL = 0 . 41 ( with corresponding evidence thresholds αH = 0 . 7925 , αL = 0 . 7650 ) . Therefore , we label as highly variable those genes for which there is strong evidence of a biological cell-to-cell heterogeneity component that explains more than 79% of the total expression variability . Similarly , we set γL = 0 . 41 , thus defining as LVG those with strong evidence that the biological cell-to-cell heterogeneity explains less than 41% of the total expression variability . Posterior estimates of the detection probabilities associated to each gene are displayed in Fig 7 . While LVG are typically associated with large expression rates , the expression rates of HVG are concentrated in a lower range . With these variance contributions and evidence thresholds , we detect 133 HVG and 589 LVG ( highlighted in Fig 7 , panels ( a ) and ( b ) , respectively ) . Among the 133 genes classified as HVG , there is an enrichment of genes related to cell differentiation ( see Table S2 in S5 Text ) . These HVG include ( posterior medians of σi are shown in parenthesis ) Dppa3 ( 85 . 1% ) for which [17] previously showed heterogeneous expression in mouse ES cells via in situ hybridisation . Other genes for which [17] found heterogeneous expression did not pass our criteria , yet we estimate a substantial component of biological cell-to-cell heterogeneity associated to most of them: Esrrb ( 79 . 7% ) , Zfp42 ( 75 . 9% ) , Krt8 ( 67 . 8% ) , Nanog ( 66 . 4% ) , Atf4 ( 64 . 6% ) , Whsc2 ( 56 . 3% ) , Rest ( 48 . 7% ) , Fscn1 ( 47 . 5% ) and Pa2g4 ( 27 . 5% ) . In particular , some of these genes would be classified as HVG if a slightly less conservative EFDR and EFNR threshold was adopted . Our results are more conservative than those according to the method described in [5] ( see Fig 8 ( a ) ) , where 1 , 363 genes were labelled as HVG . This is not surprising as their method suggests stronger heterogeneity among the cell-specific normalising constants sj’s , potentially inducing spurious heterogeneous expression in genes that remain otherwise stable . In addition , as shown in Fig 8 ( b ) , there is a relatively good correspondence between our results and the list of HVG published by [16] ( their heuristic method classified as HVG those with substantially larger expression variability than would be predicted by a Poisson model , arguing that the need of normalisation and quantification of unexplained technical noise is removed by the use of UMIs ) . There are 23 genes classified as HVG by both methods ( also detected by [5] ) , including e . g . Sprr2b ( 91 . 6% ) , Dqx1 ( 90 . 9% ) Ccdc48 ( 90 . 9% ) , Mreg ( 89 . 1% ) and Fst ( 88 . 1% ) . Several of the genes presented as HVG by [16] but not by us are borderline according to our criteria and exhibit a substantial , yet less predominant , intra-tissue heterogeneity ( the posterior medians of σi are above 68% for 75% of them ) . For example , Lefty1 exhibits a heterogeneous pattern of expression , which BASiCS reflects by estimating σi = 79 . 4% ( yet the data does not provide enough evidence to conclude that more than γH = 79% of the expression variability of Lefty1 can be attributed to a biological cell-to-cell heterogeneity component ) . Nonetheless , other genes identified as highly variable by [16] , such as Gapdh ( 35 . 7% ) , are far from being labelled as HVG by our method ( π i H ( 0 . 79 ) = 0 . 009 ) . The latter is more reasonable , in view of the extensive use of Gapdh as a reference gene in mouse ESCs [18] . The enrichment of lowly and mildly expressed genes within those highlighted as highly variable is not an artefact of our method and relates to the characteristics of the analysed mouse ESC dataset . In fact , the analysed sample includes cells from a fairly homogeneous population of cells and highly expressed genes are mostly related to key processes that are common to all cells , acting as housekeeping genes . To validate this , we analysed the dataset described in [19] , which contains 3 , 005 samples from a highly heterogeneous population of cells . In such a setting , our analysis reveals that BASiCS is capable of detecting highly variable genes across the whole range of expression levels ( see S8 Text ) . In terms of LVG , neither [5] nor [16] can be employed . Our results are validated by a strong enrichment of genes related to core cellular processes such as translation and translational elongation ( see Table S2 in S5 Text ) . In particular , we include Eif5b ( 12 . 7% ) which has been previously shown to have homogeneous expression in mouse ESCs [17] . Our list of LVG also includes e . g . : Mir466d ( 4 . 0% ) , Hsp90ab1 ( 5 . 8% ) , Gm6251 ( 11 . 4% ) , Zfp207 ( 13% ) and Arpc1b ( 14 . 0% ) . Gapdh is not labelled as LVG , however the posterior distribution of its associated σi is heavily skewed towards small values and it would be included in the LVG list if we used a slightly higher EFDR and EFNR threshold . As well as enabling estimation of the degree of unobserved technical noise , the spike-in genes can also be used to validate our method . We performed a cross-validation-type procedure where , for each of the 46 ERCC spike-in genes in turn , we modified the dataset by treating the selected technical gene as if it were a biological one . As the number of added mRNA molecules of these technical genes is known from experimental information , this experiment allows an assessment of our estimates of gene-specific normalised expression levels ( μi ) . As shown in Fig 9 ( a ) , our estimates are gathered around the true values , except for lowly expressed technical genes , where the experiment suggests a small positive bias . Estimates according to [5] are highly correlated with the true values , however the true scale has not been recovered ( not surprising as their method is not designed to estimate the right scale of the sj’s , see Fig 3 ) . In addition , we can use this analysis to validate posterior inference on the cell-to-cell biological heterogeneity components ( δi ) and our detection criteria for HVG and LVG ( see Fig 9 , panels ( b ) , ( c ) and ( d ) ) . As expected , none of the spike-in genes were detected as HVG by our criteria . On the other hand , there is strong evidence that 21 ( out of 46 ) spike-in genes fall in the LVG category , with some others just failing to overcome our conservative criteria .
Single-cell measurements of gene expression can expose heterogeneous behaviour within seemingly homogeneous populations of cells [2] . BASiCS incorporates an integrated normalisation method where cell-specific normalising constants are estimated as model parameters . In particular , we normalise expression counts according to the mRNA content of each cell . These so-called size factors are biologically important since they can partially reflect cell cycle stage ( cells tend to contain more mRNA molecules in later stages of the cell cycle [20] ) . To demonstrate this idea , we analysed the mouse ESC dataset described in [20] , where the cell cycle stage of the analysed cells was recorded . BASiCS estimates substantially larger mRNA content for those cells captured during G2 and M phases ( with respect to those in earlier stages of the cell cycle ) . A summary of this analysis is displayed in S6 Text . Additionally , our joint model of biological and spike-in genes allows biological cell-to-cell variability to be teased apart from other technical sources of variability as well as facilitating the generation of a calibrated decision rule , based on easily interpretable posterior probabilities , for selecting highly or lowly variable genes in the population of cells under study . Such information can uncover sub-populations of cells with distinct patterns of gene expression as well as producing a natural ranking of genes according to their biologically variability . Among others , future extensions of BASiCS include the implementation of differential expression analyses . BASiCS also provides a basis to build more complex downstream analyses including cluster analyses and spatial models , among others . In addition , fast advances in technology suggest that the number of sequenced cells will dramatically increase in the near future ( e . g . the one described in [19] ) , hence we foresee that a parallel implementation of the algorithm ( e . g . using graphical processing units ) might be necessary to cope with such large datasets more efficiently . | Gene expression signatures have historically been used to generate molecular fingerprints that characterise distinct tissues . Moreover , by interrogating these molecular signatures it has been possible to understand how a tissue’s function is regulated at the molecular level . However , even between cells from a seemingly homogeneous tissue sample , there exists substantial heterogeneity in gene expression levels . These differences might correspond to novel subtypes or to transient states linked , for example , to the cell cycle . Single-cell RNA-sequencing , where the transcriptomes of individual cells are profiled using next generation sequencing , provides a method for identifying genes that show more variation across cells than expected by chance , which might be characteristic of such populations . However , single-cell RNA-sequencing is subject to a high degree of technical noise , making it necessary to account for this to robustly identify such genes . To this end , we use a fully Bayesian approach that jointly models extrinsic spike-in molecules with genes from the cells of interest allowing better identification of such genes than previously described computational strategies . We validate our approach using data from mouse Embryonic Stem Cells . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | BASiCS: Bayesian Analysis of Single-Cell Sequencing Data |
Most yeast ribosomal protein genes are duplicated and their characterization has led to hypotheses regarding the existence of specialized ribosomes with different subunit composition or specifically-tailored functions . In yeast , ribosomal protein genes are generally duplicated and evidence has emerged that paralogs might have specific roles . Unlike yeast , most mammalian ribosomal proteins are thought to be encoded by a single gene copy , raising the possibility that heterogenous populations of ribosomes are unique to yeast . Here , we examine the roles of the mammalian Rpl22 , finding that Rpl22−/− mice have only subtle phenotypes with no significant translation defects . We find that in the Rpl22−/− mouse there is a compensatory increase in Rpl22-like1 ( Rpl22l1 ) expression and incorporation into ribosomes . Consistent with the hypothesis that either ribosomal protein can support translation , knockdown of Rpl22l1 impairs growth of cells lacking Rpl22 . Mechanistically , Rpl22 regulates Rpl22l1 directly by binding to an internal hairpin structure and repressing its expression . We propose that ribosome specificity may exist in mammals , providing evidence that one ribosomal protein can influence composition of the ribosome by regulating its own paralog .
Protein synthesis is a major energy consuming process involving intricate coordination of translation machinery in response to nutrient availability and stress sensing signals , as well as hormonal and growth factor cues in multi-cellular organisms . The ribosome is comprised of two ribonucleoprotein subunits: the 40S and 60S ( ‘small’ and ‘large’ subunits , respectively ) . Together these subunits facilitate peptide bond formation , performing different roles during translation . Ribosome synthesis is a highly controlled process , whereby three distinct RNA polymerases are synchronously coordinated to produce equimolar amounts of four rRNAs and 79 mammalian ribosomal proteins ( RPs ) [1]–[4] . A growing number of human diseases have been linked to mutations in genes encoding factors involved in ribosome biogenesis and protein synthesis [5] , [6] . These include developmental malformations , inherited bone marrow failure syndromes and cancer in a variety of organisms [5] , [7]–[9] . In addition , interventions leading to reduced translation , such as dietary restriction and reduced 60S ribosomal protein expression , elicits lifespan extension in yeast , worms and files [10]–[13] . Determining the molecular pathology underlying diseases and the role of ribosomes in aging requires a better understanding of ribosome specificity and the functions of individual RPs . RPs are generally thought to be essential components of the functional ribosome and although they do not play a direct role in catalyzing peptidyl transfer , they may be critical for both regulatory and structural functions of the ribosome [14] , [15] . In addition to their role in the ribosome , many RPs , including murine Rpl22 , have been shown to have extra-ribosomal functions [16]–[18] . In particular , as RNA binding proteins , RPs have been found to bind cellular and viral RNAs outside of the context of the ribosome . Some RPs also function to regulate their own expression , such as Rpl30 in yeast [19] , [20] and human RPS13 [21] . RPs are often essential for viability . For example , embryonic lethality was reported in the first murine knockout of a ribosomal protein ( RP ) gene , Rps19 [22] . Rpl24 [23] , [24] , and Rps6 [25]–[27] , also play essential roles . However , two reports have found that mice lacking either Rpl22 or Rpl29 survive without these RP genes [28] , [29] . In yeast , approximately 85% of the RP genes are duplicated as a result of an ancient genome duplication event [30] and many of these paralogous genes are functionally redundant [31] . Generally , deletion of either paralog , but not both simultaneously , results in viability; however , yeast that lack non-duplicated RP genes or both paralogs of an individual subunit are often , but not always , inviable [32] . Cross-complementation studies in yeast , analyzing defects in growth , have shown that most RP paralogs are functionally redundant [31]; however , several recent studies suggest that some paralogs might have subtle functional differences [33]–[38] . In the case of RPL22 , however , tetrad analysis indicated that the rpl22aΔ rpl22bΔ double mutant was viable , although slow growing [37] , [39] , while in worms disruption of rpl22 expression is lethal ( http://www . shigen . nig . ac . jp/c . elegans/index . jsp ) . Rpl22 is an external protein on the 60S ribosomal subunit that is incorporated into the ribosome at later stages of ribosome maturation [40] . An early study suggested that Rpl22 was not required for translation in vitro [41]; however , the protein has been identified as a component of the ribosome [40] and likely plays a role in protein translation . In addition , other activities have been attributed to Rpl22 in mammals , including association with both viral RNAs , like EBER1 , and cellular RNAs , such as human telomerase RNA [42]–[44] . Mice lacking Rpl22 are viable but have a defect in T cell development attributed to p53-dependent arrest of the αβ lineage T cells [28] . Recently , RPL22 has been found to be mutated or downregulated in various cancers , including T-acute lymphoblastic leukemias [45] , invasive breast carcinoma [46] , and lung adenocarcinoma [47] . Here we report evidence that in mice one ribosomal protein can control composition of the ribosome by regulating expression of its own paralog . Knocking out Rpl22 results in up-regulation of Rpl22-like1 ( Rpl22l1 ) , a paralog of Rpl22 whose predicted protein sequence is highly homologous to Rpl22 . Rpl22l1 was first identified ( although mis-labeled Rpl22 ) in a screen for 14-3-3ε binding partners in mouse brain [48] and has been identified as a trace component of ribosomes in mouse liver and mammary gland tissues [49] . We find that Rpl22l1 co-sediments with actively translating ribosomes in Rpl22−/− mice and a compensatory increase in Rpl22l1 expression likely accounts for the lack of translational defects in these animals . Enhanced Rpl22l1 expression also occurs upon acute knockdown of Rpl22 expression , indicating that Rpl22 has an active role in suppressing the synthesis of its paralog . Mechanistically , we find that Rpl22 directly represses expression of Rpl22l1 mRNA by binding to an internal hairpin structure . shRNA-mediated knockdown of Rpl22l1 causes a severe growth defect in cells lacking Rpl22 . Accordingly , we demonstrate that the composition of the ribosome is regulated by the novel mechanism of direct repression of one paralog by another , and offer the hypothesis that this is one mechanism by which ribosome specificity is coordinated .
A gene-trapped mouse embryonic stem cell clone harboring a mutation in Rpl22 was obtained from Bay Genomics and used to generate Rpl22 heterozygous mice ( Rpl22+/− ) ( see Text S1 ) . 5' rapid amplification of cDNA ends ( 5′ RACE ) followed by automated DNA sequencing determined that the gene-trap vector inserted between the third and fourth exons of Rpl22 ( Figure S1A ) . Gene-trap vector disruption of Rpl22 expression was confirmed by PCR and western blot analysis ( Figure S1B ) . Mice heterozygous for the Rpl22 mutation ( Rpl22+/− ) were interbred to obtain homozygous Rpl22-null ( Rpl22−/− ) mice . Surprisingly , Mendelian ratios of Rpl22+/+ , Rpl22+/− , and Rpl22−/− were found in the resulting progeny . During the construction of our mouse line , Anderson et al . reported the generation of viable Rpl22−/− mice and observed defects in lymphocyte development [28] . Characterization of our Rpl22−/− mice indicated that they have defects in lymphocyte development ( Figure S2 , S3 , S4 ) similar to that described by Anderson et al . ( 2007 ) [28] . B220+ B cells in the bone marrow were also significantly reduced in Rpl22−/− mice ( Figure S4A , B ) . Further analysis indicated that B cell development was interrupted by Rpl22 deficiency , as evidenced by a decrease in the B220+ developing IgM-IgD- and immature IgM+IgD- B cells ( Figure S4C , D ) . Despite the ubiquitous expression of Rpl22 and its hypothesized role in mRNA translation , disruption of Rpl22 in mice results in a remarkably mild phenotype . Hematologic parameters are normal in these mice [50] . Also , unlike deletion of RPL22 in yeast [51] , Rpl22−/− mice have no substantial difference in growth rate or size relative to Rpl22+/+ and Rpl22+/− littermates ( Anderson et al . 2007 and our unpublished data ) . Surprisingly , no significant differences were observed in the polysome profiles of lysate from Rpl22−/− liver , lung or cultured ear fibroblasts when compared to samples collected from Rpl22+/+ mice ( unpublished data ) , indicating that Rpl22 is not essential for translation efficiency or ribosome biogenesis in the tissues evaluated . We considered the possibility that another factor might be compensating for lack of Rpl22 in mice . A bioinformatic search identified Rpl22-like1 ( Rpl22l1 ) as a candidate . Rpl22l1 encodes a 122 amino acid protein that is 73% identical to Rpl22 ( Figure 1A ) . The protein sequence of Rpl22l1 is highly conserved from human to zebrafish ( Figure S5 ) . To determine if significant levels of Rpl22l1 mRNA exists in tissues from Rpl22+/+ mice and whether Rpl22l1 transcript levels increase in Rpl22−/− mice , lung , liver , spleen and kidney were harvested from Rpl22−/− mice and their littermate controls and analyzed by quantitative RT-PCR ( qRT-PCR ) for Rpl22 and Rpl22l1 expression with Acidic Ribosomal Protein ( ARBP ) mRNA levels used for normalization ( Figure 1B ) . In Rpl22+/+ samples , high Rpl22 expression was detected , while Rpl22l1 transcripts were less abundant . In samples isolated from Rpl22−/− mice , qRT-PCR revealed a ∼3-fold induction of Rpl22l1 mRNA expression relative to littermate controls . Rpl22l1 transcripts were found associated with actively translating ribosomes in both Rpl22+/+ and Rpl22−/− mouse ear fibroblast samples ( unpublished data ) , suggesting that the Rpl22l1 mRNA is actively translated into protein . Consistently , equivalent increases in Rpl22l1 protein levels are observed in the absence of Rpl22 in lung , liver , spleen and kidney ( Figure 1C ) . Similar increases in Rpl22l1 mRNA and protein expression were observed in Rpl22−/− skeletal muscle and brain ( unpublished data ) . Additionally , Rpl22l1 was found to be abundantly expressed in both Rpl22+/+ and Rpl22−/− pancreas ( Figure 1C ) . These results indicate that Rpl22 negatively regulates , either directly or indirectly , Rpl22l1 expression in a range of mouse tissues . To determine if Rpl22l1 is incorporated into actively translating ribosomes , liver tissue was isolated from Rpl22−/− mice and their littermate controls followed by sedimentation of the lysates on sucrose gradients . Fractions collected from the gradients , were subsequently loaded onto an SDS-page gel for western blot analysis . In Rpl22−/− samples , Rpl22l1 is present in fractions containing 60S ribosome subunits and polysomes , suggesting that it is incorporated into free ribosome subunits and ribosomes actively translating mRNA in the absence of Rpl22 ( Figure 2 ) . Rpl7 , a RP that is incorporated into the large subunit of the ribosome , is present in the fractions containing 60S ribosome subunits and polysomes in both Rpl22+/+ and Rpl22−/− samples . Rpl22 and Rpl22l1 , but not Rpl7 , were detected in fractions 1 and 2 , representative of the free , non-ribosomal lysate ( Figure 2C , D ) , consistent with the hypothesis that these RPs exist in states within the cell both associated with the ribosome and independent of the ribosome . Additionally , while Rpl22l1 levels are relatively evenly detected in 60S containing fractions of the polysome profile ( Figure 2D , fractions 3–7 ) Rpl22 is detected at higher levels in the fractions containing free 60S subunits and messages loaded with fewer ribosomes ( Figure 2C , fractions 3–5 vs 6–7 ) . To verify that Rpl22l1 was incorporated into 60S subunits and actively translating ribosomes , free 60S subunits and 80S monosomes from actively translating polysomes were isolated from liver lysates of Rpl22+/+ ( WT ) and Rpl22−/− ( KO ) mice using sucrose density gradient fractionation ( Figure 2E , F; see Text S1 ) . Samples were then concentrated and prepared for mass spectrometry analysis using standard methods ( see Text S1 ) . In order to measure the relative amounts of Rpl22 and Rpl22l1 in the WT and KO mouse liver lysate samples , we used multiple reaction monitoring ( MRM ) , a targeted mass spectrometry ( MS ) approach that is highly sensitive . MRM is a targeted type of MS and requires a list of peptide targets and their subsequent fragment targets—known as a transition list—to program the analysis on the instrument ( see Text S1 ) . The final MRM analysis consisted of 4 peptide targets for Rpl22 ( 5 counting uniquely modified targets ) and 3 peptide targets for Rpl22l1 , and was limited to the top 8 fragment ions per peptide , creating a total of 64 transitions targeted in the analysis . Summing the integrated MRM peak areas for all transitions from all observed peptides for each protein ( Rpl22 and Rpl22l1 ) yielded the total MRM peak areas plotted for each of the four tested samples ( WT60S , WT80S , KO60S , KO80S ) and indicated the relative amounts of Rpl22 and Rpl22l1 in these samples ( Figure 2G , H ) . These data indicate that Rpl22l1 levels are significantly higher in the 60S and 80S subunits isolated from Rpl22−/− liver than in Rpl22+/+ littermate controls ( Figure 2H ) , supporting the hypothesis that Rpl22 regulates Rpl22l1 expression and , as a result , incorporation into ribosomes . Why does expression of Rpl22l1 increase in mouse tissues lacking Rpl22 ? We considered two potential explanations: ( 1 ) Rpl22 directly regulates Rpl22l1 expression , or ( 2 ) compensation occurs during development in Rpl22−/− mice . To distinguish between these two possibilities , Rpl22 was acutely knocked down in 3T9 fibroblasts using a lentiviral-mediated inducible knockdown system that allows doxycycline-inducible regulation of Rpl22 and changes in Rpl22l1 expression were examined . 3T9 cells were transduced with 2 different tet-on shRNA lentivirus constructs ( shRNA 1 and shRNA 2 ) that target Rpl22 mRNA or a nonspecific control construct . Following 3 days of doxycycline treatment , Rpl22l1 mRNA expression is enhanced 1 . 8 fold in 3T9 cells with reduced Rpl22 expression ( Figure 3A ) . Western blot analysis confirmed that Rpl22l1 protein levels were elevated by the knockdown of Rpl22 , while expression of other RPs , such as Rpl7 remained unchanged ( Figure 3B ) . These results confirm that Rpl22 negatively regulates the expression of Rpl22l1 acutely and raise the possibility that Rpl22-mediated regulation of Rpl22l1 is an active process with biological significance . Collectively , these data suggest that Rpl22 is regulating expression of Rpl22l1 but the mechanism leading to the increased expression is unknown . To determine whether Rpl22 affects the stability of Rpl22l1 mRNA , cultures of 3T9 cells were treated with actinomycin D , which blocks transcription by all three eukaryotic polymerases [52] . After actinomycin D treatment , the levels of Rpl22l1 mRNA in Rpl22+/+ 3T9 cells decreased significantly ( p<0 . 01 ) relative to the untreated control , while in Rpl22−/− 3T9 cells Rpl22l1 levels were maintained and the rate of decay was reduced ( Figure 4A ) . These results suggest that Rpl22 affects the stability of Rpl22l1 mRNA . If Rpl22 is involved in mediating Rpl22l1 stability , does Rpl22 bind directly to Rpl22l1 mRNA ? Previous studies determined that Rpl22 is associated with Epstein-Barr virus-expressed RNA , EBER1 [43] , [53] and evaluation of the RNA binding specificity of Rpl22 suggested that Rpl22 recognizes a stem loop ( hairpin ) structure with a G-C at the neck followed by a U [42] , [53] . To address whether regulation of Rpl22l1 expression is directly mediated by Rpl22 , an algorithm termed M-fold that predicts RNA secondary structure was used to evaluate Rpl22l1 mRNA structure for potential Rpl22 RNA binding motifs [54] . Analysis revealed the presence of a consensus Rpl22 RNA-binding motif within exon 2 of zRpl22l1 , suggesting that Rpl22 might interact directly with Rpl22l1 mRNA ( Figure 4B ) . To test whether Rpl22 can directly bind Rpl22l1 mRNA via the hairpin structure identified in Rpl22L1 mRNA by M-fold analysis an RNAse protection analysis was performed . Recombinant proteins were incubated with radiolabeled RNA , and UV-crosslinked . The RNAs were then digested with RNase A , and samples were run on a protein gel . Proteins bound to radiolabeled RNA were detected by autoradiogram at their expected molecular weight . Unbound proteins are not detected on autoradiogram . Recombinant Rpl22 was found to bind to in vitro transcribed zRpl22L1 mRNA , but not the zRpl22l1 mRNA lacking the hairpin structure ( zRpl22L1Δhp ) ( Figure 4C ) , suggesting the Rpl22 directly binds to Rpl22l1 mRNA . To determine if Rpl22 directly regulates expression of Rpl22l1 , we employed a biosensor quantification assay using GFP as a fluorescent indicator of effects on expression . Zebrafish embryos were microinjected with mRNAs EGFP-zRpl22 , EGFP-Rpl22l1 or a mutant form of Rpl22l1 in which the Rpl22l1 hairpin was modified ( EGFP-zRpl22L1mt ) in combination with constructs expressing zRpl22 or zRpl22l1 . mCherry mRNA was co-injected to allow for quantification of the relative fluorescence intensity . Co-injection of zRpl22 with EGFP-Rpl22l1 led to a significant decreased fluorescence relative to those embryos co-injected with zRpl22 and EGFP-Rpl22l1mt ( Figure 5A–E ) , suggesting that the hairpin structure within Rpl22l1 mRNA is necessary for Rpl22 to directly regulate its expression . Next , to assess if the presence of the hairpin structure within the Rpl22l1 mRNA is sufficient to regulate expression , the hairpin sequence from zRpl22l1 mRNA was fused to EGFP and evaluated in the biosensor quantification assay . The heterologous reporter mRNA , zRpl22l1-150h-EGFP , containing the minimal sequence identified by M-fold to form the hairpin structure , was co-injected with mCherry mRNA ( injection control ) and ( Figure 5G ) zRpl22 mRNA or ( Figure 5I ) Rpl22-Morpholino ( Rpl22-MO ) into zebrafish embryos . Rpl22 repressed the expression of zRpl22l1-150h-EGFP reporter , while knockdown Rpl22 can increase the expression of reporter , suggesting the hairpin structure identified in zRpl22l1 mRNA is sufficient to regulate mRNA abundance . In yeast , ribosomal protein paralogs are thought to functionally compensate for one another , each incorporating into the ribosome in the absence of the other . Deletion of both yeast RPL22 paralogs results in viable , but slow growing cells [37] . Recently Rpl22+/− or Rpl22−/− primary mouse embryonic fibroblasts ( MEFs ) were found to grow faster and display increased transformation potential relative to MEFs isolated from Rpl22+/+ littermates [45] To further evaluate the effect of Rpl22 and Rpl22Ll1 expression on growth rates , Rpl22+/+ or Rpl22−/− 3T9 fibroblasts were transduced with one of 2 different tet-on shRNA lentivirus constructs ( shRNA#1- and shRNA#2-Rpl22l1 ) that target Rpl22l1 mRNA to acutely knock-down its expression . Western blot analysis confirmed that Rpl22l1 protein levels were elevated in Rpl22−/− 3T9 fibroblasts ( Figure 6A ) . No significant difference was observed in the growth rates of Rpl22+/+ or Rpl22−/− 3T9 fibroblasts ( Figure 6B ) . In doxycycline-treated Rpl22+/+ or Rpl22−/− 3T9 fibroblasts expressing the shRNA constructs , Rpl22l1 protein levels were confirmed to be reduced by western blot analysis ( Figure 6C , E ) . Knockdown of Rpl22l1 significantly reduced growth rates of Rpl22+/+ fibroblasts and greatly impaired that of Rpl22−/− 3T9 fibroblasts ( Figure 6D , F and S6 ) , indicating that cells lacking both paralogs have severe growth defects . In contrast , acute knockdown of Rpl22 in Rpl22+/+ 3T9 fibroblasts resulted in no change in the rate of proliferation ( Figure S7 ) . In summary , these cell culture studies indicate that expression of at least one paralog of Rpl22 is required for normal growth and suggests that Rpl22l1 may also affect cell proliferation by a mechanism independent of Rpl22 .
Here we report on phenotypes of mice lacking the large subunit ribosomal protein , Rpl22 . Surprisingly , Rpl22−/− mice have no developmental defects , other than previously reported defects in T and B cell development ( Anderson et al . 2007 and our data ) . More generally , Rpl22−/− mice have no significant defects in translation , as judged by sucrose gradient sedimentation to detect ribosome occupancy on transcripts ( unpublished data ) . This is somewhat surprising since yeast lacking Rpl22 function are slow growing with substantial defects in translation [10] , [37] . One possible explanation is that the lack of a defect in translation in Rpl22−/− mice is explained by compensatory increases in expression of another gene , Rpl22l1 , which shares a high degree of homology with Rpl22 . While Rpl22l1 mRNA is detected at low levels in most tissues of wild-type mice , mRNA and protein levels are dramatically increased in the Rpl22 knockout . It is interesting that Rpl22l1 is expressed at relatively high levels in the pancreas even though Rpl22 is present as well . Future studies will be needed to determine why these two paralogs are jointly expressed in this tissue . When expressed , Rpl22l1 was incorporated into ribosomes and actively translating polysomes . These findings indicate that Rpl22l1 is capable of functioning within ribosomes that are actively translating mRNA . Future studies examining the function of Rpl22l1 will help decipher to what extent Rpl22 and Rpl22l1 may have redundant roles and also determine their independent functions , for which accumulating evidence is emerging [28] , [45] ( O'Leary et al , unpublished data ) . Although poorly understood , gene compensation during development is a recurrent phenomenon in mouse knockout studies [55]; therefore , enhanced expression of Rpl22l1 might reflect developmental compensation in Rpl22−/− mice . Alternatively , Rpl22 could play an active role in the repression of Rpl22l1 . To test this , we examined the consequences of acute knockdown of Rpl22 in 3T9 fibroblasts . Our findings indicate that Rpl22l1 mRNA and protein levels are rapidly increased following Rpl22 knockdown and support the conclusion that developmental compensation does not account for the increased Rpl22l1 mRNA and protein levels . Instead , we find that inhibition of transcription in cells lacking Rpl22 results a slower decay of Rpl22l1 mRNA compared to Rpl22+/+ cells , suggesting Rpl22 affects the stability of Rpl22l1 mRNA . It is possible that more than one mechanism is involved in increasing Rpl22l1 expression in the absence of Rpl22 . Further studies interrogating other mechanisms are needed to full understand what regulates Rpl22l1 upon Rpl22 deficiency . We find that Rpl22 binds to a hairpin structure in the Rpl22l1 mRNA . The hairpin motif identified in Rpl22l1 mRNA is necessary and sufficient for regulation of mRNA abundance by Rpl22 . These data suggest that Rpl22 might function in an extra-ribosomal capacity to bind and destabilize Rpl22l1 mRNA . Rpl22l1 is yet another RNA demonstrated to interact with Rpl22 . The viral RNA EBER1 contains three Rpl22 binding sites and is thought to compete with the 28S rRNA for Rpl22 binding in Epstein-Barr virus-infected cells [42] , [43] , [53] , [56] , [57] . In yeast , recent studies have revealed regulation of the expression of one paralog in a duplicated RP gene pair by the other . Ribosomal protein S29A ( Rps29a ) , for example , regulates its own expression along with expression of its paralog , RPS29B [58] . Interestingly , Rpl22 has also been ascribed functions independent of its role as a component of the ribosome [17] , [18]; these extra-ribosomal functions include regulation of telomerase activity [44] and association with histone H1 [59] . In fact , many ribosomal proteins , including Rpl7 , Rpl13a , and Rps3 [60]–[62] have extra-ribosomal functions , which will have to be considered as mechanistic links are sought to explain diseases associated with RP mutations and possibly age-related phenotypes . Given their ancient nature , it is not surprising that evolution has settled on ways of exploiting these proteins for multiple uses . Do Rpl22 and Rpl22l1 have shared or unique functions ? The observation that one ribosomal protein represses expression of its own paralog is likely to be of biological significance and determining whether mice lacking Rpl22l1 have specific phenotypes will identify tissues or conditions where Rpl22l1 is the functional RPL22 paralog that participates in translation . We propose that Rpl22 and Rpl22l1 share the ability to participate in protein translation as part of the ribosome , since both paralogs are found incorporated into ribosomes and Rpl22 has an active role in suppressing expression of Rpl22l1 . However , unique roles are also well established , since knockout of mouse Rpl22 leads to specific phenotypes in T and B cell development ( Anderson et al . 2007 and our data ) , and data from Zhang et al . indicate that both Rpl22 and Rpl22l1 are essential for T cell development in zebrafish and exhibit antagonistic functions in regulating the emergence of hematopoietic stem cells [63] . Together , these observations lead us to propose that Rpl22 and Rpl22l1 may have overlapping activities , sharing a role in enhancing large subunit ribosome function but also having distinct roles in development of the immune system [63]and , in the case of Rpl22l1 , perhaps other tissues . Based on data from S . cerevisiae , in which genes for most ribosomal proteins are duplicated , Komili et al . proposed the existence of a ribosome code , whereby ribosome subunits composed of different ribosomal proteins would have differential specificity for mRNAs [38] . This proposal was based on an accumulation of data , including differential localization of paralogs and large-scale phenotypic screens , which indicated that specific subsets of ribosomal protein genes were often identified in phenotypic screens of gene deletion strains . This is an exciting hypothesis that , if corroborated , would identify ribosome composition as a new mechanism for regulation of gene expression . Our studies in yeast raise the possibility that ribosome specificity may occur in yeast aging [10] , [37] . In a long-lived , slow-growing and translation-compromised rpl22aΔ background , we find that deletion of the other paralog , RPL22B , causes no significant further reduction in translation , but blocks lifespan extension ( Steffen et al . , unpublished data ) . One possible explanation for this result is that the portion of ribosomes containing Rpl22b is increased in the rpl22aΔ background , leading to specific changes in translation conducive to enhanced replicative lifespan . Other explanations are possible and further studies will be necessary to test this important hypothesis . Structural studies do not point to an obvious mechanism by which Rpl22 paralogs would influence the pool of translated RNAs and other models involving specific non-ribosomal functions of Rpl22 and Rpl22l1 have to be identified and/or tested . One limitation to the ribosome specificity hypothesis proposed in S . cerevisiae is that ribosomal proteins are generally not thought to be duplicated in other organisms . Therefore , if ribosome specificity indeed exists [38] , it may be more prominent in yeast than mammals . Recently , Xue and Barna have suggested that specialized ribosomes might also regulate gene expression in mammals [64] . Our findings support the hypothesis , at least in the case of murine Rpl22 , that one ribosomal protein may repress expression of the other , raising the possibility that ribosome specificity may extend to organisms other than yeast . In addition , a recent study of Rpl38 mutant embryos found that although global translation was unchanged , translation of Hox genes were altered [65] , providing additional support for the hypothesis that differential composition of the ribosome might contribute to transcript-specific translational regulation . Cryo-EM studies of the eukaryotic 80S ribosome have demonstrated that Rpl38 is located on the surface of the ribosome and interacts with a region of the rRNA known as expansion segment 27 ( ES27 ) , which has two distinct orientations toward the L1 stalk or toward the exit tunnel [66] , [67] . The location of Rpl38 in the ribosome is consistent with its proposed role in regulating transcript-specific translation . The tissue specific defects observed in Rpl22−/− and Rpl29−/− mice , along with mice expressing mutated Rpl38 [65] , suggest that temporal and spatial expression of RPs are critical for proper development and tissue patterning . In plants and Drosophila , many RPs paralogs display tissue specific variations and are differentially expressed during development [68]–[71] . In mammals , recent studies found mRNA expression patterns of RPs vary in different tissues and cell types [65] , [72] . Together with the findings presented here , these studies illustrate that heterogeneous expression of RPs is tightly regulated; however additional studies will be crucial in confirming the influence of these differential expression patterns on specialized ribosome activity and message specific translation . Further studies are also required to address the questions that have arisen from this study . Are there tissues or cell types in mice where Rpl22l1 and not Rpl22 is the predominant paralog ? What is the mechanism by which Rpl22 represses expression of Rpl22l1 through interaction with its mRNA ? What are the non-ribosomal functions of Rpl22 paralogs ? And most importantly , do ribosomes with Rpl22l1 have different specificity for mRNAs than those with Rpl22 ? Findings presented in this study provide interesting leads in which to test the hypothesis that specialized ribosomes exist and , furthermore , point to a new level of regulation in ribosome biogenesis , wherein ribosomal paralogs regulate each other's synthesis to optimally maintain the organism .
A detailed description of the targeting vector along with the generation and genotyping of the Rpl22−/− mice can be found in the Text S1 . DMEM tissue culture medium was purchased from Mediatech ( Manassas , VA ) , L-glutamine , penicillin/streptomycin , and trypsin were from GIBCO ( Carlsbad , CA ) . Fetal calf serum was from Gemini ( West Sacramento , CA ) . Rpl22 antibody was from BD Biosciences ( Franklin Lakes , NJ ) , Rpl22l1 antibody was from Santa Cruz ( Santa Cruz , CA ) and the Rpl7 antibody was purchased from Novus Biologicals ( Littleton , CO ) . GAPDH antibody was from Ambion ( Austin , TX ) . HRP-conjugated donkey anti-rabbit and donkey anti-mouse were from Jackson Immunoresearch Laboratories , Inc ( West Grove , PA ) . Rpl22 , Rpl22L1 , and ARBP primers were from Operon ( Huntsville , AL ) . Lentiviral shRNA constructs V2MM_120192 and V2LHS_131608 , directed at Rpl22 ( denoted shRNA 1 and 2 , respectively ) , along with V3LMM_473587 and V3LHS_322499 , directed at Rpl22l1 ( denoted shRNA 1 and 2 , respectively ) , were purchased through Open Biosystems ( Huntsville , AL ) and . A non-silencing-TRIPZ lentiviral inducible shRNAmir control was also purchased through Open Biosystems . RNA was isolated from mouse tissue or 3T9 cells using the RNeasy kit ( Qiagen ) with a DNaseI treatment following the manufacturer's protocol . cDNA synthesis was performed using oligo ( dT ) and reverse transcriptase Superscript Preamplification System ( Invitrogen ) . Quantitative PCR was performed using a Biorad iCycler and measuring SYBR green incorporation for product detection . The fold-increase of Rpl22 and Rpl22l1 were normalized to the housekeeping gene ARBP . The primer sequences are as follows: RPL22 ( forward ) : 5′-GTCGCCAACAGCAAAGAGAG-3′; RPL22 ( reverse ) : 5′-TCCTCGTCTTCCTCCTCCTC-3′; RPL22l1 ( forward ) : 5′-TGGAGGTTTCATTTGGACCTTAC-3′; RPL22l1 ( reverse ) : 5′-TTTCCAGTTTTTCCATTGACTTTAAC-3′ . For qRT-PCR analysis of gradient fractions , values for Rpl22 and Rpl22l1 were normalized to an artificial polyadenylated RNA that was added in equal amounts to each gradient fraction . 3T9 cells were treated with 1 µM Actinomycin D for the indicated time ( 0 , 2 , 4 , 6 , 8 or 24 hours ) . Cells were washed once with PBS then total RNA was isolated and cDNA was synthesized as described above . Relative quantitation was calculated by the 2−ΔΔCT method and normalized to ARBP . Adult mouse tissue or cultured cells were lysed in RIPA buffer ( 300 mM NaCl , 1 . 0% NP-40 , 0 . 5% sodium deoxycholate , 0 . 1% SDS , 50 mM Tris pH 8 . 0 ) and proteins separated and transferred to nitrocellulose membranes using the Invitrogen ( Carlsbad , CA ) Nu-Page system . Proteins in polysome fractions from sucrose density gradients ( described above ) were TCA precipitated and resuspended in 30 ul of Sample Buffer ( 1× Laemmli buffer with bromophenol blue and DTT ) , while for mouse liver polysome fractions , 32 ul of each 1 ml fraction was loaded and proteins were separated and transferred as described above . Blots were blocked with 5% milk in TBST for 1 h and incubated overnight at 4°C with the following dilutions of primary antibodies: Rpl22 ( 1∶250 ) , Rpl22l1 ( 1∶250 ) , Rpl7 ( 1∶1000 ) , GAPDH ( 1∶50 , 000 ) . Membranes were incubated for 2 h with HRP-conjugated donkey anti-rabbit or sheep anti-mouse ( 1∶10 , 000 ) secondary antibodies and antigen was detected using enhanced chemiluminescence ( ECL Plus Western Blotting Detection System , Amersham ) . Polysome analysis of mouse liver samples was adapted from the protocol of Oliver et al . [23] . Minced liver tissue was homogenized on ice using a Dounce homogenizer in ice-cold lysis buffer containing 1 . 5 mM KCl , 2 . 5 mM MgCl2 , 5 mM Tris-HCl , pH 7 . 4 , 1% Na deoxycholate and 1% Triton X-100 . After centrifugation at 2500 g for 15 min at 4C in a microfuge , heparin and cycloheximide were added to the clarified supernatants to final concentrations of 1 mg/ml and 100 µg/ml , respectively , and centrifuged at 16 , 000 g for 10 min at 4°C . Samples containing 20–25 OD260 units in 1 ml were loaded onto 11 ml 7–47% ( w/v ) sucrose high-salt ( 800 mM KCl ) gradients described previously [73] . Samples were centrifuged in a Beckman SW40 Ti rotor at 39 , 000 rpm ( 270 , 500×g ) for 2 hr at 4°C and 1-ml fractions were collected from the top of the gradients with a Brandel fractionator system . The A254 gradient profiles shown in Figure 2 were digitized using a DATAQ DI-148U data recording module that converts and exports analog absorbance readings to analysis software . A detailed description of the sample preparation and the experimental procedure can be found in the Text S1 . Embryos were harvested at day 13 . 5dpc and mouse embryonic fibroblasts ( MEFs ) were harvested as previously described [74] . Following isolation , MEFs were cultured in DMEM ( Dulbecco's modified Eagle medium ) containing 10% FBS , 1% Penicillin/Streptomycin and 1% L-Glutamine and subjected to 3T9 ( Figure 6 ) protocol [74] , [75] . Cells were removed from tissue culture plates with 0 . 05% trypsin and 0 . 02% EDTA for 3 min at 37°C , washed with DMEM , and plated overnight prior to experimentation . For Rpl22 and Rpl22l1 knockdown , lentiviral pGIPZ shRNA constructs were purchased from Open Biosystems and cloned into inducible pTRIPZ constructs . Briefly , 293T cells were transfected with pTRIPZ constructs bearing shRNA against Rpl22 ( shRNA 1 and 2 ) or a non-silencing control using calcium phosphate transfection . After 48 h , viral supernatant was harvested and used to transduce 3T9 cells . Cells were selected for 5 days in puromycin ( 5 µg/ml ) followed by 3 days of doxycycline ( 1 µg/ml ) treatment to induce shRNA expression before experimentation . For cell proliferation assays , Rpl22+/+ or Rpl22−/− 3T9 cells were infected with control shRNA or 1 of 2 shRNA constructs targeting Rpl22 or Rpl22l1 . After 5 days of selection in puromycin , growth rates of cells transduced with each shRNA construct were determined by plating cells in triplicate at a density of 30 , 000 cells/well in 6-well plate in media with or without doxycycline ( 1 µg/ml ) . Every 3 days for the cells were counted and replated at 30 , 000 cells/well for 15 days . Protein was isolated from 10 cm plates treated in parallel to the growth assays . Protein sequences for murine Rpl22 ( NP_033105 ) and Rpl22l1 ( NP_080793 ) were obtained from Genbank ( http://www . ncbi . nlm . nih . gov/ ) . Sequence alignments were performed using ClustalW2 Algorithm ( http://www . ebi . ac . uk/Tools/msa/clustalw2/ ) . mRNA sequences for mRpl22l1 ( NM_026517 ) and zRpl22l1 ( NM_001045335 ) were entered into mFold ( http://mfold . rna . albany . edu/ ? q=mfold/RNA-Folding-Form ) for prediction of RNA secondary structure . Statistical comparisons were made between groups by a 2-tail Mann-Whitney U test ( Figure 6 , S5 , S6 ) , by Newman-Keuls post hoc analysis , a two-tailed Student's t test ( Figures 4 , S2 , S3 , S4 ) , or a 1-way ANOVA ( Figure 3 ) . Significant differences between groups are indicated in figure legends . AB wild-type zebrafish strain was bred and maintained at 28 . 5°C under standard aquaculture conditions . Embryos were staged as described previously [76] . Full-length coding sequence for EGFP , mCherry , zebrafish Rpl22 ( zRpl22 ) and Rpl22l1 ( zRpl22l1 ) were cloned into pCS2+ . Full- length cDNA sequences encoding EGFP-zRpl22 and EGFP-zRpl22l1 were subcloned in the pCS2+ . The zRpl22l1mut , mutant form of zRpl22l1 in which hairpin GATGGGATTCTCGATT was mutated to GACGGTATCTTAGACT , was generated using GeneTailor kit ( Invitrogen , Carlsbad , CA ) with the modified forward primer 5′-GACTGCACTCACCCTGTGGAGGACGGTA TCTTAGACTCT GCAAACTTTG-3′ . The 150 bp fragment in the zebrafish Rpl22l1 CDS containing the Rpl22-binding hairpin was amplified by PCR using the following primers: Forward: cagggatccatgcagactgttgtgagaaagaat , Reverse : tcagaattccaggttgcctgttttgccattaac . Then the PCR product was digested ( BamH1+EcoR1 ) and ligated into pCS2+EGFP to get the in-frame chimera sequence , which was referred to as zRpl22l1-150h-EGFP . All mRNAs for microinjection were synthesized using the mMessage mMachine kit ( Ambion , Austin , TX ) . Then embryos were injected at one-cell stage with indicated synthetic mRNAs and observed at 10 hpf stages . Injection doses: 100 pg for each mCherry , GFP-Sensor and Inhibitor mRNAs . Morpholinos were ordered from Gene Tools ( Gene Tools , LLC , Philomath , OR ) and dissolved in nuclease-free water . Morpholino to bind the translation start site ( AUG MOs ) of zebrafish Rpl22 [63] ( Sequence: 5′-CCGACAGTTTTGGCAGAAAGCCAGT-3′ ) was injected at 6 ng in the 1-cell stage embryos . The standard control MO from Gene Tools was used as a control ( Control MO , sequence: 5′-CCTCTTACCTCAGTTACAATTTATA-3′ ) . Images were taken from the Nikon SMZ1500 stereomicroscope equipped with DS-Fi1 digital camera and Nikon Ar imaging software . Fluorescence sensor assays was designed and performed according to the well-established method in zebrafish [77] , [78] . To calculate relative fluorescence for each embryo , green or red pixel intensity was quantified using ImageJ software ( NIH ) . Relative fluorescence was determined for each embryo ( GFP/mCherry ) . After normalization to the average GFP/mCherry ratio in the control embryos , the sensor repression or enhancement fold was obtained in the inhibitor mRNA or Rpl22-MO injected groups . The coding regions of EBER1 and EBER 2 were amplified by PCR from the cDNA of EBV-infected African Burkitt lymphoma KemIII cells ( kindly provided by Jeffrey Sample , Penn State Hershey College of Medicine , Hershey , PA ) , adding a T7 promoter , and cloned into pCR2 . 1 plasmid ( Invitrogen , Carlsbad , CA ) . The primers used for PCR were: forward 5′-TAATA CGACTCACTATAGGCAAAACCTCAGGACCTACGCTG , TAATACGACTCACTATAGGTCAAA CAGGACAGCCGTTGC and reverse 5′-GAACTGCGGGATAATGGATGC , AAGCCGAATACC CTTCTCCCAG for EBER 1 and EBER2 respectively . The zRpl22l1Δhp and EGFP-zRpl22l1mut were generated from plasmid zRpl22l1-pCS2+ and EGFP-zRpl22l1-pCS2+ , respectively , with deletion of the hairpin TGGGATTCTCGA using GeneTailor mutagenesis kit according to manufacturer's instruction ( Invitrogen , Carlsbad , CA ) . pGEX-m88 plasmid was generated from pGEX-hRpl22 ( kindly provided by Joan Steitz , Yale University , New Haven , CT ) using GeneTailor mutagenesis kit according to manufacturer's instruction ( Invitrogen , Carlsbad , CA ) creating a EEYLKE motif instead of KKYLKK . The plasmids used for RPA assay were EBER 1 , EBER 2 , zRpl22l1 and zRpl22l1Δhp . The plasmids were linearized using EcoRI for EBERs and NotI for zRpl22l1s ( enzymes from New England Biolab , Ipswich , MA ) , and radioactive probes were prepared by in vitro transcription with T7 ( EBERs ) or SP6 ( zRpl22like1s ) RNA polymerases at 37°C ( Maxiscript , Ambion , TX ) , in presence of 10 µCi of 3ZP-UTP ( Perkin Elmer , Boston , MA ) , and purified by G-50 columns ( Illustra Probe-Quant , GE Healthcare , Buckinghamshire , UK ) . The radiolabeled RNAs were renatured for 2 min at 95°C and kept on ice for 5 min . Radiolabeled RNAs and protein were incubated on ice for 12 min , in presence of 5 µg of tRNA ( Roche ) in RPA buffer ( 10 mM Tris pH 8 . 0 , 50 mM NaCl , 0 . 75 mM MgCl2 , 0 . 8 mM DTT and 2 . 5% glycerol ) , and the mixtures were cross-linked on ice pack for 30 min ( using a UV Stratalinker 1800 ) . After cross-linking , the mixture were digested by RNase A ( Qiagen , Valencia , CA , 7 , 500 U/µl ) diluted to 3 , 750 U/µl in RNase A buffer ( 20 mM Tris pH 7 . 0 , 2 mM MgCl2 , and 0 . 2 mM KCl ) for 20 min at 37°C , then 2 µl of denaturing dye were added before denaturation for 2 min at 90°C . Nine µl of the reaction mixtures were loaded on 4–12% Bis-Tris gel , 1 . 5 mm ( Invitrogen , Carlsbad , CA ) . The gels were dried and expose to BioMax MR films ( Kodak , Rochester , NY ) or to a phosphoimager plate read using a Fuji BAS-2500 reader . E . coli strain BL21DE ( 3 ) pLys ( Promega , Madison , WI ) was used to produce recombinant GST , GST-hRpl22 and GST-hRpl22m88 from plasmids pGEX-3X . pGEX-hRpl22 and pGEX-hRplm88 . Isopropyl-β-D-thiogalactopyranoside ( IPTG ) at a final concentration of 100 µM was added to 2× YT media at optical density A600 = 0 . 4 , and incubation was maintained for 4 hours at 37°C . The cells were harvested and resuspended at 0 . 25 mg/ml in 1× PBS ( 2 . 6 mM KCl , 1 . 7 mM KH2PO4 , 137 mM NaCl , 11 mM Na2HPO4 , pH 7 . 4 ) and lysed by 3 passages in a M-110L Pneumatic ( Microfluidics , Newton , MA ) , in presence of Complete mini without EDTA tablets ( Roche , Mannheim , Germany ) . The cell extracts were loaded onto glutathione sepharose beads for batch purification ( GE Healthcare , Upsala , Sweden ) at 2 ml/liter of culture . The beads , 0 . 5 ml , were washed with 1 ml of wash buffer ( 50 mM Tris , pH 9 . 5 , 150 mM NaCl ) , then washed with 1 ml of bump buffer ( 50 mM Tris pH 9 . 5 , 150 mM NaCl , 1 mM reduced GSH ) . The proteins were eluted with 3 fractions of 1 ml elution buffer ( 50 mM Tris pH 9 . 5 , 150 mM NaCl , 10 mM reduced GSH ) . The protein purity was assessed by SDS gel electrophoresis , and the concentration measured using Coomassie Plus ( Bradford ) Protein Assay ( Pierce , Rockford , IL ) . The proteins were stored in 10% glycerol in 20 µl aliquots at −80°C . Appropriate protocols of all work on mice have been approved by the IACUC committees at the Institution where the work was performed and is in accord with accepted national guidelines . | Translation is the process by which proteins are made within a cell . Ribosomes are the main macromolecular complexes involved in this process . Ribosomes are composed of ribosomal RNA and ribosomal proteins . Ribosomal proteins are generally thought to be structural components of the ribosome but recent findings have suggested that they might have a regulatory function as well . A growing number of human diseases have been linked to mutations in genes encoding factors involved in ribosome biogenesis and translation . These include developmental malformations , inherited bone marrow failure syndromes and cancer in a variety of organisms . Here , we describe the role of one ribosomal protein regulating another . We provide evidence that ribosomal proteins can influence the composition of the ribosome , which we hypothesize , may impact the function of the ribosome . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"animal",
"models",
"biochemistry",
"protein",
"synthesis",
"model",
"organisms",
"proteins",
"gene",
"expression",
"genetics",
"regulatory",
"proteins",
"biology",
"mouse",
"rna",
"stability"
] | 2013 | The Ribosomal Protein Rpl22 Controls Ribosome Composition by Directly Repressing Expression of Its Own Paralog, Rpl22l1 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.